Keywords

1 Introduction

One in three learners in the German education system comes from an immigrant family, and Germany currently is the country with the largest number of new recent immigrants in Europe. In many other countries around the globe, considerable proportions of the respective populations are also immigrants and their descendants, and there is little reason to assume that migration movements will decline in the near future. The successful integration of immigrants and their children into the education systems is of major importance for these individuals’ future prospects and for the social cohesion and viability of the receiving societies. Identifying factors and processes that foster (or hinder) educational performance is therefore pivotal. Adopting an interdisciplinary perspective and drawing on NEPS data, the project ‘The role of immigrants’ first and second language proficiency for social integration, particularly in education: Analyses of NEPS data’ aimed to clarify the role of immigrants’ language proficiency and language use in their educational success.

The education of immigrants and their children is an important socio-political issue and a thriving area of research in a number of disciplines including education, sociology, and psychology. This research consistently indicates that in most contexts including Germany, immigrants and their offspring tend to be less successful in the education system than native-born ethnic majority students (for an overview, see Edele & Stanat, 2022). On average, they attain lower levels of school-related competencies (e.g., Henschel et al., 2019; Rjosk et al., 2017), with particularly large disparities in the first generation and even more pronounced disparities among recently arrived refugee students (Schipolowski et al., 2021). Students of immigrant descent are also at a disadvantage with respect to their educational participation and the degrees they eventually complete. For instance, they participate less frequently in early education (Autorengruppe Bildungsberichterstattung, 2018), and attend the academic school track (Gymnasium) less often than students from the native-born ethnic majority (Kristen & Granato, 2007; Weis et al., 2018). They also have lower chances of attaining university entrance qualifications and higher chances of leaving school without a certificate (Statistisches Bundesamt, 2019).

These findings raise the question regarding which factors and processes account for the described inequalities. One well-established and key result of previous research is that the difficulties immigrants and their children face in their educational careers are due largely to differences in their parents’ educational and social background (Gresch & Kristen, 2011; Heath et al., 2008; Henschel et al., 2019; Kristen & Granato, 2007; Müller & Stanat, 2006; Rjosk et al., 2017). They are a result of social rather than of migration-related inequalities (Kalter, 2005). However, even when students’ social background is taken into account, educational disparities remain—at least for some groups of origin and on some indicators of educational success. The remaining differences point to additional factors that are specific to the post-migration situation. One of these factors is language.

From the point of view of a general resource framework (Becker, 1962, 1993), immigrants often lack resources that are important for their children’s educational success and for their labour market integration. One prominent example for such a resource is proficiency in the language of the residence country or immigrants’ second language (L2). L2 proficiency is generally viewed as a key to integration (Chiswick & Miller, 1995; Esser, 2006; Kalter, 2011). Immigrant families frequently speak the language used in their country of origin (often labelled as first language or L1) at home—either exclusively or in addition to L2. Children growing up in immigrant families thus often have fewer opportunities to acquire their L2 in their families, and, on average, attain lower L2 competence levels than their peers from native families (for an overview, see Kempert et al., 2016).

At the same time, they have a special resource at their disposal: namely, proficiency in L1. Whereas theory, empirical findings, and practitioners agree on the crucial role of access to and knowledge of L2 for students’ educational success, the relevance of their L1 for their education is controversial (see Edele et al., 2020; Kempert et al., 2016). Although the role of L1 for the acquisition of L2 and for school success has been discussed for decades, available evidence on the consequences of being competent in L1 and using it in everyday interactions is inconclusive. One major problem is that language proficiency, especially L1 proficiency, is often measured with self-reports, even though the validity of the resulting findings is limited (see Edele et al., 2015). Indeed, studies using objective measures of language proficiency are scarce, particularly in data sets with large samples.

The data gathered in the National Educational Panel Study (NEPS; see Blossfeld et al., 2011) offer a unique opportunity to examine the role of L1 (and L2) in language-minority students’ educational success, because they include language tests that objectively assess language-minority students’ proficiency in L1. More specifically, NEPS assesses proficiency in the two most commonly used L1s in the German immigrant population—Turkish and Russian—in several starting cohorts. In addition, it includes a range of key indicators of educational outcomes, including objective tests of L2 reading comprehension, English skills, and mathematics and science proficiency. NEPS data are also rich in important background characteristics and potential confounding variables such as students’ socio-economic background or their general cognitive abilities.

The present contribution focuses on one of the project’s key research questions: the role of L1 in language-minority students’ educational success. In the following section, we describe main theoretical positions addressing this relationship. We then present key findings from our project, and conclude with a discussion of these results and of future research directions in this area.

2 The Role of L1 in Language-Minority Students’ Educational Success: Theoretical Background

Different theoretical positions suggest beneficial, absent, or detrimental consequences of L1 proficiency and use on language-minority students’ educational outcomes. In the following, we present key arguments and positions that characterize these views. We start with arguments that characterize L1 as a resource for education.

2.1 L1 as a Resource for Educational Success

Two major positions suggest that language-minority students benefit from their L1. These are the cognitive perspective and the cultural perspective (Mouw & Xie, 1999). The cognitive perspective assumes that L1 proficiency or use—either on its own or in combination with L2—fosters learners’ cognitive abilities and competence development in school-related domains. The linguistic interdependence hypothesis (Cummins, 1979, 1981, 2000) provides a key argument for this perspective. It states that language-minority learners with a high level of proficiency in their L1 have advantages in acquiring L2. Specifically, it posits that the linguistic level that learners achieve in L2 is a function of the linguistic level they have previously achieved in L1. This proposition is based on the idea that command of any language relies on a so-called common underlying proficiency (CUP) through which linguistic, metalinguistic, and conceptual knowledge is transferred across languages. The assumed positive effect of L1 proficiency on acquiring L2 should promote language-minority students’ educational success across domains.

In addition, it has been suggested that bilingualism—that is, L1 proficiency and use in combination with L2 proficiency and use—has beneficial cognitive effects (see Kempert et al., 2016). According to one prominent argument, the alternate use of different languages and the constant demand to control two language systems enhances executive functioning (Bialystok, 2017). Executive functioning is needed to play with ideas, to take time to think before acting, to master novel challenges, to stay focused, and to resist temptations (Diamond, 2013). These abilities are relevant for learning processes, and executive functioning has been shown to be an important predictor of academic success (e.g., Huizinga et al., 2018). If multilingualism does, in fact, enhance executive functioning, it should facilitate the school learning of multilinguals across learning domains.

It has also been proposed that bilingualism entails advantages for the acquisition of additional languages (L3). A core argument for this proposition is that multilinguals possess an enhanced metalinguistic awareness that should facilitate the acquisition of other languages (e.g., Thomas, 1988). Metalinguistic awareness is the ability to view language as an object and to reflect on language in abstract terms (Jessner, 2006). Multilinguals may also benefit from their enhanced executive functioning in their L3 learning, and be able to transfer knowledge and skills from the languages they already know to new languages (Hirosh & Degani, 2018). Language-minority learners are therefore often assumed to outperform their monolingual peers in acquiring third languages.

The cultural perspective also proposes positive effects of L1 on language-minority students’ educational success, but it introduces a different mechanism. It assumes that L1 use can provide access to important resources of L1-speaking family members and close contacts in the co-ethnic community. Proponents of the theory of segmented assimilation argue that, under certain circumstances, close ties to family members and to individuals in the co-ethnic community can protect children and adolescents from adverse influences (Bankston & Zhou, 1995; Mouw & Xie, 1999; Portes & Hao, 2002; Zhou, 1997). Family members and co-ethnics can provide social support and may also exert social control and promote educational success as an important goal. In this way, close-knit networks can prevent students from socializing in subcultures that counteract their educational progress. Note that this position suggests positive effects of L1 only under certain conditions—namely, when learners grow up in contexts in which they are at risk of entering deviant subcultures, and when family members and other close co-ethnic ties provide the necessary resources and orientations (see Kroneberg, 2008). Moreover, scholars have pointed out the transitional character of these influences: being able to communicate in L1 should be helpful only as long as immigrant students’ close contacts have limited command of L2. With better knowledge of L2 in students’ co-ethnic environments, these advantages should disappear (Mouw & Xie, 1999; Portes & Hao, 2002).

The outlined perspectives concurrently suggest beneficial effects of L1 for language-minority students’ educational success, yet, they differ in essential aspects. The cognitive perspective suggests benefits of high levels of L1 proficiency and of alternating use of L1 and L2 for individuals’ competence development in different educational domains. According to the cultural perspective, the use of L1 in everyday communication is the key to accessing important resources that family members and co-ethnics have at their disposal. Although the use of L1 requires at least a basic level of proficiency, the crucial factor here is the communication rather than the achieved skill level or the switching between languages. The positive consequences of using L1 are expected to extend to a range of outcomes including competence development as well as educational decisions. The cognitive perspective, in contrast, focuses exclusively on the consequences of L1 on competence development.

2.2 L1 as Irrelevant or Detrimental for Educational Success

Other scholars argue that L1 is irrelevant for immigrants’ educational success. According to their reasoning, integration in the receiving society relies primarily on resources that are immediately relevant and can be used directly in this context such as the respective majority language(s) (Esser 2004, 2006, 2009; Kalter, 2007). Ethnic resources, such as L1, are considered to be of little usefulness in a receiving context in which they can rarely be employed. Accordingly, this position suggests that most L1s are of neglectable relevance in educational institutions, unless they are taught as foreign languages or used as a language of instruction, which is generally not the case—at least in the German context. For this reason, language-minority children should not benefit from their L1 and should not have advantages in their educational outcomes compared to their monolingual peers. L1 is thus seen as largely irrelevant for language-minority learners’ education.

The so-called time-on-task hypothesis assumes a competing relationship between the use of L1 and L2 (e.g., Gathercole, 2002; Leseman et al., 2009; Scheele et al., 2010). Time on task refers to the time that learners actively spend on what they need to learn (Hopf, 2005). The idea that time on task matters for learning outcomes goes back to Carroll’s (1963, 1973) learning theory according to which learning success is a function of the time that learners actively invest and the time they need to acquire a certain learning content. This position emphasizes that learners have only a limited amount of learning time at their disposal, and that the time spent on acquiring their L1 takes away from the time that could be allocated to acquiring L2 and other school-relevant competencies. Accordingly, the time spent on using L1 at home should have negative effects on the acquisition of L2 and, ultimately, on educational success. Table 16.1 provides an overview of the theoretical perspectives suggesting various effects of L1 on educational success.

Table 16.1 Overview on theoretical perspectives suggesting effects of L1 on educational success and project findings
Full size table

3 The Role of L1 in Language-Minority Students’ Educational Success: Key Findings

The various empirical studies conducted within the project ‘The role of immigrants’ first and second language proficiency for social integration, particularly in education: Analyses of NEPS data’ provide multifaceted insights into the role of L1 in language-minority students’ educational success. Analyses were based on cross-sectional data, carefully controlling for relevant potentially confounding factors such as the socio-economic and sociocultural background of the family, general cognitive abilities, or type of school attended. First, we present findings pertaining to the theoretical perspectives that view L1 as a resource, starting with the cognitive perspective followed by the cultural perspective. Then, we report results on language competition and thus on arguments that characterize L1 as an obstacle to immigrant students’ educational success.

3.1 Interdependence Between L1 Listening Comprehension and L2 Reading Comprehension

To elucidate the role of L1 in L2 reading—a key indicator of educational success and a crucial prerequisite for succeeding in school and later in life—we determined how language-minority students’ listening comprehension in L1 relates to their reading comprehension in L2 (Edele & Stanat, 2016). Reading models, including the prominent simple view of reading (Gough & Tunmer, 1986; Hoover & Gough, 1990), regard listening comprehension as a major determinant of reading comprehension. Previous research has confirmed the importance of listening comprehension for reading comprehension, particularly in older students (Carver, 1998; Kendeou et al., 2009; Tilstra et al., 2009). At the same time, research on the linguistic interdependence hypothesis (Cummins, 1979, 2000) has focused mainly on the effect of L1 literacy skills and its precursors on L2 reading abilities (e.g., Lindsey et al., 2003; Prevoo et al., 2016; Proctor et al., 2006). Examinations of the transfer of L1 oral language skills such as listening comprehension to reading in L2, in contrast, are scarce (see Geva & Genesee, 2006).

To address this gap, we used data from NEPS Starting Cohort 4Footnote 1 that assessed language-minority students’ listening comprehension in Russian and Turkish as well as their L2 reading comprehension (for details on test construction and validation, see Edele et al., 2016; Gehrer et al., 2013). We tested the assumption that L1 listening comprehension is related to language-minority students’ reading comprehension in L2, and that this linguistic interdependence is more pronounced at higher levels of L1 proficiency. The sample included 502 ninth-grade students with Russian as L1 and 662 ninth-grade students with Turkish as L1 from the first, second, and third immigrant generation. Russian and Turkish differ in their similarity to students’ L2, with Russian being more similar to German than Turkish. Nevertheless, in both language groups, L1 listening comprehension significantly predicted L2 reading comprehension in linear regression models. This was also true after important control variables were taken into account, including students’ socio-economic background, L2 vocabulary knowledge, L2 reading fluency, and their general cognitive abilities. Polynomial regression models further indicated that the relationship between L1 proficiency and L2 proficiency was linear in the Russian sample, yet stronger at higher levels of L1 proficiency in the Turkish sample. Overall, these findings support the assumption of linguistic transfer and indicate that oral language skills transfer as well. They also partly confirm the idea that transfer is more pronounced at higher levels of L1 proficiency. A possible explanation for the differential findings across the two language groups is the different degree of similarity to German of either Turkish or Russian. In distant languages, such as Turkish and German, only language-independent skills such as metacognitive strategies have the potential to transfer. The transfer of such skills, however, may be restricted to advanced levels of L1 listening comprehension. More similar languages such as Russian and German, in contrast, allow for the transfer of language-specific skills such as vocabulary knowledge. This should also be possible at lower proficiency levels.

In demonstrating that L1 listening comprehension relates positively to L2 reading comprehension, the study extends the range of L1 skills previously known to transfer to L2 reading. It indicates that cross-linguistic transfer is not restricted to literacy skills, but extends to oral language abilities. This finding is highly relevant from a practical perspective. Language-minority students typically acquire their L1 in their families. They may therefore speak their L1 fluently but not necessarily be literate in this language. Our findings suggest that immigrant students do not necessarily need literacy skills to benefit from proficiency in their family language. Rather, L1 oral language skills seem to have the potential to improve L2 reading comprehension. Our findings are also important from a theoretical perspective, because the linguistic interdependence hypothesis posits the transfer of literacy skills (Cummins, 2000, p. 173), and previous research on cross-linguistic transfer has focused on reading skills. Our findings suggest that theoretical accounts addressing linguistic transfer should also include the oral domain. However, L1 was associated only weakly with L2 in our study. Presumably, we would have detected a closer link between L1 and L2 if we were to have assessed reading comprehension in both languages.

3.2 Immigrant Bilingualism and Third Language Learning

The cognitive perspective also posits that bilingualism—that is, combined proficiency in L1 and L2—entails cognitive advantages. In particular, bilingualism is often assumed to benefit the acquisition of an L3. Previous research does, in fact, indicate that bilingualism is helpful for L3 learning in bilingual school contexts (for overviews, see Cenoz, 2003; Hirosh & Degani, 2018). However, only a few studies have investigated the role of bilingualism in the L3 learning of immigrant students in monolingual contexts in which only L2 is used in instruction at school and L1 is acquired in the family context (e.g., Hesse et al., 2008; Maluch et al., 2015; van Gelderen et al., 2003). Findings from these studies are inconclusive. One possible explanation for these inconsistent results is that L3 advantages depend on immigrant students’ proficiency levels in L1 and L2—something that previous studies have rarely considered.

We addressed this issue by analysing data from NEPS Starting Cohort 4Footnote 2 (see Edele et al., 2018). Assessments in this starting cohort include psychometrically sound proficiency tests of L1 (Russian/Turkish), L2 (German), and L3 (English), and allow us to determine the role of different proficiency levels in L1 and L2 in the third-language learning of language-minority students in a monolingual school context. Specifically, we compared the L3 proficiency of language-minority students with varying proficiency levels in L1 and L2 and resulting profiles of bilingualism to monolingual students’ L3 proficiency. Based on the literature, we expected only students with high proficiency levels in L1 and L2 (balanced bilinguals at a high level) to outperform their monolingual peers. To ensure that potential L3 advantages of bilinguals are not merely the result of high proficiency in the language of instruction, we compared bilinguals with different profiles not only to the entire sample of monolinguals with, by definition, average skills in German, but also to a subsample with above-average German proficiency levels. To determine the respective relevance of L1 and L2 in L3 learning for bilingual students, we carried out additional analyses focusing on this group and including L1 and L2 proficiency as continuous variables.

We analysed data from 8752 tenth graders and distinguished Russian–German (N = 352) and Turkish–German (N = 436) bilingual students from German monolingual students (N = 7964). Students’ listening comprehension test scores in Russian and Turkish served as measures of their L1 proficiency; reading comprehension test scores in German as measures of their L2 proficiency; and reading comprehension test scores in English as measures of their L3 proficiency. Multiple linear regression analyses that controlled for sociodemographic and socio-economic background characteristics, general cognitive abilities, and the type of school attended revealed that Russian–German and Turkish–German high-level balanced bilinguals outperformed their German monolingual peers with average German proficiency in the English reading test. Comparisons of high-level balanced bilingual students to their monolingual peers with similarly high levels of L2 proficiency, however, did not reveal L3 advantages of bilingual students. Additional analyses focusing exclusively on bilinguals showed that only L2 proficiency but not L1 proficiency was positively related to students’ L3 proficiency. Overall, our findings indicate that bilingualism is not necessarily beneficial for L3 learning. In a context such as Germany in which L2 is the dominant language of instruction, advantages in L3 learning seem to be limited to bilinguals with above-average proficiency levels in their L1 and L2, and occur only in comparison to monolingual students with average proficiency in L2 but not to monolingual students with above-average L2 proficiency. Moreover, only bilingual students’ proficiency in L2, but not in L1, was related to their L3 achievement. Hence, the most relevant linguistic resource for L3 learning of bilingual immigrant students in monolingual contexts seems to be their proficiency in the language of instruction (L2).

3.3 L1 Use and Mathematics Achievement

In addition to analyses of potential cognitive mechanisms associated with L1 proficiency, we investigated whether speaking L1 might entail advantages for language-minority students’ educational success due to a communicative mechanism, as suggested by the cultural perspective. According to the theory of segmented assimilation, language-minority students can gain access to favourable resources provided within the family or ethnic community by using their L1 in everyday communication (see Section ‘L1 as a resource for educational success’). Previous research examining this communicative mechanism has been inconclusive (Kristen & Olczyk, 2013). One major limitation of most empirical investigations is that they examined effects of L1 proficiency—more precisely, of self-reported proficiency in L1—rather than the use of L1 (e.g., Bankston & Zhou, 1995; Lutz & Crist, 2009; Waters et al., 2010), and were therefore unable to disentangle communicative from cognitive mechanisms. The few studies that explicitly examined L1 use did not yield conclusive evidence (Abada & Tenkorang, 2009; Dollmann & Kristen, 2010; Mouw & Xie, 1999; Xie & Greenman, 2011).

Using data from 1662 ninth-grade language-minority students of different origins (Turkey, Poland, Former Soviet Union) participating in NEPS Starting Cohort 4,Footnote 3 we investigated the association between language-minority students’ L1 use with parents and their mathematics achievement (Strobel, 2016). Adopting a conditional perspective, we expected positive effects of using L1 within the family on educational achievement only if parents hold attitudes and resources in favour of educational success. The findings from linear regression analyses revealed a negative relationship between L1 use with parents and students’ mathematics achievement. Students who spoke exclusively German at home outperformed those who also use the L1 with parents. Interestingly, this negative association was reduced to non-significant levels when considering students’ L2 reading proficiency. Thus, for students with similar proficiency levels in German, L1 use with parents does not seem to entail advantages or disadvantages. To test the assumption of a conditional effect, we further examined whether the relationships between L1 use and mathematics achievement varied with parents’ educational aspirations, their highest educational level, and ethnic origin. However, the results did not change in any of these specifications, indicating that L1 use with parents is equally (un)important.

These findings challenge the notion that using L1 within the family facilitates language-minority students’ educational achievement due to a communicative mechanism. However, we cannot rule out the possibility that using L1 might be more important under conditions and for outcomes other than the ones examined in our research. For instance, the communicative mechanism could primarily foster the transmission and reinforcement of shared values and aspirations in favour of education (e.g., Feliciano & Lanuza, 2016; Friberg, 2019) that should be particularly relevant for students’ education choices, but may be less so for their achievement.

3.4 L1 Exposure and L1/L2 Proficiency

As described above, other theoretical notions view L1 as detrimental or, at best, irrelevant for educational success (see Section ‘L1 as irrelevant or detrimental for educational success’). In line with the time-on-task hypothesis, prior studies (e.g., Azzolini et al., 2012; Duursma et al., 2007; Leseman et al., 2009; Müller & Stanat, 2006; Scheele et al., 2010), including those using data from large-scale school assessments (e.g., Haag et al., 2016; OECD, 2020), have consistently identified a negative relationship between L1 use in the family and language-minority students’ proficiency in L2 and other competence domains. These findings are in line with the proposition that using a language is one of the most powerful determinants of second-language acquisition (Chiswick & Miller, 2001; Kristen, 2019). Language-minority students who speak mainly their L1 at home rather than L2 are less exposed to the language of instruction, and this might impede L2 learning and ultimately their educational achievement.

Several analyses conducted within the framework of our project provide insights into how language exposure relates to language proficiency. In line with prior research, we found that the language used within the family related strongly to L2 proficiency. Frequently using L2 in the family context seems to increase language-minority students’ chances of achieving high levels of L2 reading abilities (Miyamoto et al., 2020; Strobel & Seuring, 2016). At the same time, we found that regularly speaking L1 at home related positively to students’ proficiency in this language (Kristen et al., 2019; Seuring et al., 2020), negatively to their proficiency in L2, and vice versa (Edele & Stanat, 2016). Moreover, we found a similar—although less pronounced and less consistent—pattern for the use of L1 with classmates and language proficiency in L1 and L2. Regularly speaking L1 with classmates related negatively to language-minority students’ L2 proficiency and positively to Turkish–German speaking students’ proficiency in L1. In the Russian–German language group, the use of L1 with classmates was unrelated to students’ L1 proficiency (Seuring et al., 2020). Taken together, these findings indicate a positive association between the use of L1/L2 and students’ proficiency in the respective language, and a negative association between L1 use and L2 proficiency. These findings are generally in line with the notion of time on task. Table 16.1 summarizes the project findings.

4 Conclusions

Our research conducted within the project ‘The role of immigrants’ first and second language proficiency for social integration, particularly in education: Analyses of NEPS data’ provides multifaceted insights into the role of L1 proficiency and L1 use in language-minority students’ educational attainment. A key finding is that students with good L1 oral skills have greater chances of attaining high levels of L2 reading skills. This finding is in line with the linguistic interdependence hypothesis and complements previous research that has focused mainly on the cross-linguistic transfer of reading abilities. Relating to the consequences of bilingualism for third-language learning in the monolingual school context of Germany, our analyses do not support the notion that immigrant bilinguals are generally at an advantage. Only bilinguals with above-average proficiency levels in both L1 and L2 outperform monolingual students with average proficiency in L2. Moreover, only bilingual students’ L2 proficiency, but not their L1 proficiency, relates positively to their L3 proficiency. This suggests that the language of instruction, rather than bilingualism, is the main linguistic resource for acquiring a third language in the monolingual school context of Germany. Taken together, these findings partly support the cognitive perspective. At the same time, they demonstrate that positive effects of L1 are neither very strong nor consistent.

The notion that using the L1 opens access to parental resources that support language-minority children’s educational success was not supported in our analyses. Specifically, we did not find evidence supporting the cultural perspective according to which speaking the L1 facilitates educational success because of a communicative advantage, as suggested by the theory of segmented assimilation. This is in line with previous studies in the German context (e.g., Dollmann & Kristen, 2010) that did not find benefits of using L1 at home for language-minority students’ educational success. Overall, the pattern of our findings implies that the L1 can be a potential resource for language-minority students’ educational success due to cognitive processes, but not because of an enhanced communication with other speakers of L1.

Our findings also provide insights into the role of language exposure in language-minority students’ L1 and L2 acquisition. The frequent use of a language, particularly in the family context and to a lesser extent with classmates, relates positively to proficiency in this language and negatively to proficiency in the respective other language. These findings support the time-on-task hypothesis and are also in line with theoretical models of second language acquisition (e.g., Chiswick & Miller, 2001) according to which language exposure is of key relevance for language learning. At the same time, it is still unclear to what extent and under what conditions negative effects of using L1 on L2 learning due to time on task are exacerbated or even outweighed by positive effects of bilingualism due to enhanced cognitive functioning and transfer.

Despite the negative net influence of L1 use on L2 proficiency, we would like to emphasize that we do not support the call for language minority families to use L2 instead of L1 at home in order to support their children’s educational success. This would not just ignore their individual rights and deprive them of the opportunity to transmit their L1 to their children. It might also jeopardize their children’s educational integration, especially if parents have only limited proficiency in L2 and cannot provide high-quality input in this language. At the same time, findings suggest that it is important to provide language-minority children with sufficient and high-quality exposure to L2 in the education system, and to start this already in preschool. This includes L2 input from different sources, including native speakers, that is adequately challenging in its complexity (Unsworth, 2016).

One important limitation of our analyses is their cross-sectional design. Although we carefully controlled for potentially confounded variables, it is not possible to draw causal conclusions or to identify changes over time. Future research should and will exploit the potential of the NEPS for longitudinal analyses, keeping in mind that it could take longer periods of time for potential effects of L1 and bilingualism to unfold. Another promising venue for future research is to examine the role of immigrant bilingualism on school adjustment more generally and ask whether bilingualism entails advantages for competence development in additional domains such as mathematics and science as well as in cross-domain skills such as metacognition. The interrelations of L1 proficiency and L1 usage with other resources—for instance, the identification of students of immigrant descent with their ethnic group and the ethnic majority group, and their psychological adaptation (Schotte et al., 2018)—are also promising venues for future work in this field.

Overall, there is no simple answer to the question which role L1 plays in language-minority children’s educational success. The question is complex and needs to be answered in a differentiated fashion. First, it is necessary to specify whether the question relates to L1 use or L1 proficiency. Second, it is necessary to take proficiency levels in L1 and L2 into account. Third, effects vary for different educational outcomes.

To conclude, we would like to stress that speaking their L1 is an important resource for students from immigrant families, regardless of potential effects on their educational outcomes. Speaking another language, and particularly the language of one’s ethnic origin, is valuable in its own right. It allows one to communicate with speakers of this language, to connect to the cultural heritage of one’s ethnic group and participate in its cultural life, and, generally speaking, to gain experiences one would have missed without these skills. It is important to keep this in mind when interpreting findings on effects of L1 and, most importantly, when drawing conclusions about potential implications for policy and practice.