Introduction

Recently, the number of children with foreign roots has been increasing in the settings of early childhood education and care and developmental support. Their families' national origins and first language are also diversifying (e-stat, 2019). These children grow up in a bilingual environment and speak their first language (i.e., heritage language^*1) and Japanese. Many of their parents cannot communicate fluently in Japanese. Therefore, childcare and social workers face difficulties communicating with these children and parents. Childcare and social workers need to have an appropriate understanding of the developmental status of children. However, in the case of bilingual children, one of the biggest challenges is understanding the actual situation of their language development.

In most cases, monolingual children have well-balanced skills between verbal comprehension and verbal expression. However, in the case of bilingual children, there might be a large developmental gap between verbal comprehension and expression skills. A lot of bilingual children can understand what others say but nevertheless cannot express themselves effectively. In addition, the learning speeds of Japanese and that of the heritage language are not the same. Children are more likely to learn and use a language spoken at home. Such a developmental gap between two languages or different language skills brings about difficulty in evaluating the actual status of language development in bilingual children.

Another factor that hinders the accurate evaluation of language skills in bilingual children is the use of reference values obtained from standardized tests designed for monolingual children. It is inappropriate to use such values to calculate the developmental ages and indexes for bilingual children. Needless to say, if children whose mother tongue is not Japanese receive assessments in Japanese, it will never be possible to capture their potential. As Shimoida (2014) pointed out, such standard values should be treated just as references. Therefore, it is difficult to obtain specific figures such as developmental ages and quotients for bilingual children. The factors of cultural differences and parental socioeconomic status also hinder the accurate understanding of language development in bilingual children.

Nevertheless, to identify and support the potential of preschool children with foreign roots, the assessment of language development is critical. Otherwise, they may not receive necessary help at the early childhood stage and may fail to fully develop their abilities at school due to the language barrier. Furthermore, according to a previous study (Uccelli and Páez, 2007), the vocabulary and narrative^*2 skills of preschool children in their heritage language could affect the future progress of language learning at school. Therefore, in my current research study, I seek a methodology to evaluate the status of language development, in particular the vocabulary development of young children who grow up in a bilingual environment. I am also pursuing studies to ascertain whether reading aloud to children in the heritage language can improve children's ability to retell the story in Japanese. In this article, I will discuss some results of my research.

Studies on children's vocabulary development

First, I will introduce some studies on the assessment of vocabulary development in children. According to previous studies conducted by researchers in Europe and the US, children who were raised in a bilingual environment (in particular, those from families of lower socioeconomic classes) were more likely to exhibit delayed language development (Bialystock et al., 2010; Uccelli & Páez, 2007). Some researchers in English-speaking countries proposed the use of a measure of conceptual vocabulary to evaluate vocabulary development in children who grow up in a bilingual environment (Pearson et al., 1993). The theory of a conceptual vocabulary is that, if a child receives a vocabulary test in two languages and answers correctly in at least one language, the child is deemed to have the concept of the word. By using this assessment method, underestimating children's potential to acquire vocabulary knowledge can be avoided (Core et al., 2013).

However, there are certain issues to consider when applying a measure of conceptual vocabulary to bilingual children using Japanese and another language. For example, the Japanese word "miru" has several corresponding words in English, such as "look," "watch," and "see." Or one may even wonder if there is another appropriate translation. It seems that there are not many words that have exactly the same concept in two languages. In particular, highly abstract words such as adjectives and verbs are more difficult to match the concept in two languages compared to nouns that refer to concrete objects. In my current research project, the theory of conceptual vocabulary to assess children's vocabulary development has been employed. After considering the above shortcoming, I determined to conduct a vocabulary test in Japanese and Portuguese^*3 using nouns with specific meanings.

[Survey Participants]
The survey participants are 30 children from a five-year-old class and a six-year-old-class of a company's in-house daycare center located in an area with a high concentration of Brazilian residents. In this article, however, I only present the data of 10 children (from five-year and two-month-old to six-year and 10-month-old; average age: six-year and two-month-old; five boys and five girls; three from five-year-old class, and seven from six-year-old class), for which the analysis was completed. All of their parents use Portuguese. In the daycare center, two childcare workers speaking Japanese and one speaking Portuguese take care of the class. Childcare activities are conducted in Japanese. In addition, a Portuguese lesson is provided for six-year-old children for about one hour every day. Children speak Portuguese when interacting with other peers. Overall, children are living in a Portuguese-dominant environment.

[Assessment]
For vocabulary assessment, I used a Brazilian Standard Expressive Vocabulary Test (Teste de Vocabulário Expressivo) consisting of 100 questions asking the names of pictures shown to children (Capovilla et al., 2011). The Portuguese-speaking staff conducted the vocabulary test in Portuguese, and the Japanese-speaking staff conducted the vocabulary test in Japanese, respectively.

[Results]
The test results were analyzed by classifying them into the following categories: "Wrong answers/no answers in both languages," "Correct answers in Japanese only," "Correct answers in Portuguese only," and "Correct answers in both languages." Correct answers in at least one language were counted as "conceptual vocabulary." The average number of wrong answers/no answers was 22.9, while the average number of correct answers in both languages was 32.5. The average number of correct answers in Japanese (the sum of correct answers in Japanese and those in both languages) was 41.2. The average number of correct answers in Portuguese (the total of correct answers in Portuguese and those in both languages) was 68.4. Finally, the average number of conceptual vocabulary items was 77.1 (Table 1). These results revealed that the score of the conceptual vocabulary assessment was higher than the score of the one-language assessment, indicating children's higher expressive vocabulary skills.

Table 1: Average number of correct answers to 100 questions (n=10)

It might be quite predictable that the score of the two-language assessment was higher than the score of the one-language assessment. Nevertheless, the findings in this survey indicate the risk of underestimating the potential language abilities of children with foreign roots when conducting only a one-language assessment. Although there are some shortcomings in a two-language assessment, it is important to seek a way to understand the actual status of bilingual children's potential in language development. To achieve this, various means can be used, including the employment of specialists and interpreters who can speak foreign languages (Shimoida, 2014; Sasaki, 2014), interviews with parents about children's family life, and the use of translation software.

In addition, the above results revealed that the participant children achieved higher assessment scores in Portuguese than in Japanese. This trend is more apparent among six-year-old children (detailed figures are omitted in this article), which is attributed to the fact that the participant children are living in a Portuguese-dominant environment. I recognize that, for young children, the heritage language spoken at home is critical for building a solid parent-child relationship. This also ensures their language development and identity formation after entering school. In this regard, it is necessary to provide early childhood education and care by valuing the importance of Portuguese as a heritage language. At the same time, support should be provided to enhance a Japanese environment necessary for their social life. To do so, we first need to understand the developmental gap between the two languages by conducting a two-language assessment.

Studies on language assessments using a story-retelling task

According to a survey on Spanish-English bilingual children, their narrative skills in Spanish (their dominant language in early childhood) will affect the quality of narrative skills in English after they enter school (Uccelli & Páez, 2007). Similarly, children whose parents are Brazilian residents in Japan may be able to improve their narrative skills in Japanese by leveraging their dominant language, Portuguese.

Here, I will explain my exploratory study on bilingual children who have difficulties with narrative discourse in Japanese. In this study, I used a story-retelling method with a picture book, which seems to be a relatively easy task for children. Children were asked to retell a story that had been read to them aloud. Further studies are needed to find out whether their ability to retell a story can improve their narrative skills to describe their own experiences. In my opinion, a story-retelling task, in which children are required to convey a narrative structure with the passage of time, could enhance their narrative skills.

However, for children who have difficulty understanding a picture book in Japanese, it is not practical to read them a story in Japanese only. They may be unable to completely understand and enjoy the story. Instead, we should read them a story in both Japanese and their heritage language. This will provide the children with an opportunity to comprehensively leverage their language skills.

They can learn new vocabulary items, understand the story characters and their emotional states, and exercise their imagination to follow the narrative structure. In addition, bilingual children can use their comprehension of the story in their heritage language to understand the story in Japanese, which may improve their Japanese skills. Therefore, I started my research study based on the assumption that telling a story in two languages can improve children's ability to retell the story in Japanese.

[Methodology]
Using a story-retelling task, a survey was conducted targeting the above 10 participant children (aged five to six years old). I divided these children into two groups according to their chronological age, gender, and the level of Japanese receptive and expressive vocabulary. For the first group of children, a story was read aloud in Portuguese first and then in Japanese. For the second group of children, the story was read aloud twice in Japanese. The picture book used in this survey is published in both Portuguese and Japanese, containing eight old tales (Cousins, 2009). The story chosen for this survey was one out of these eight stories, which is about a bear family and a little girl, in 16 pages, intended for children aged four to six. After hearing the story read aloud, the participant children were asked to retell the story in Japanese, and then, in Portuguese.

After the survey, the recordings of children's story-retelling were transcribed into text data. Then, a quantitative analysis of the data was conducted according to the total number of utterances, the mean length of utterances (MLU), and the total number of different words (TDW, meaning the number of new core vocabulary words) in Japanese and Portuguese.

[Results]
At this point, the results of the Japanese analysis on five children of each group have been obtained. For the Portuguese analysis, the work is currently in progress.

When comparing the first group (telling a story in two languages) and the second group (telling a story twice in Japanese), the second group achieved higher scores than the first group. However, considering the limited number of participants, it might be too early to reach a conclusion on the outcome of story-retelling in Japanese. At this stage, the results are quite predictable, showing that telling a story twice in Japanese is more effective than telling a story one time in Japanese and Portuguese to inspire children's story-retelling ability. Analysis on the Portuguese data of story-retelling will be continued to find out more about the effect of story-telling in two languages.

The vocabulary data of the above ten children were also analyzed to find any correlation between their Japanese/Portuguese vocabulary and Japanese MLU/TDW. As a result, it is confirmed that a stronger correlation was found between Portuguese vocabulary and Japanese MLU/TDW. Therefore, although it is too early to reach a conclusion until sufficient data analysis is completed, children's vocabulary in their heritage language may affect their ability to retell a story in Japanese. From this perspective, it can be said that if children's acquisition of their heritage language is given sufficient assistance in early childhood, this will positively affect their acquisition of the Japanese language.

ACKNOWLEDGEMENT
I would like to express my sincere appreciation to all children, parents, and daycare caregivers for their kind support and cooperation in my survey. This research project received the Grants-in-Aid for Scientific Research, Basic Research B (No. 17H02718), by the Japan Society for the Promotion of Science (JSPS).

• *1: The term "heritage language" means a language learned from parents and grandparents except local languages.
• *2: The term "narrative" means the making of a "story" about events and experiences conveying the passage of time and emotions.
• *3: The term "Portuguese language" used in this survey means Portuguese that is the official language of Brazil.

References

• Bialystock, E., Luk, G., Peets, K.F. & Yang, S. (2010). Receptive vocabulary differences in monolingual and bilingual children, Bilingualism: Language and Cognition, 13(4), 525-531.
• Capovilla, F.C., Negrão, V.B., & Damázio, M. (2011). Teste de Vocabulário Auditivo e Teste de Vocabulário Expressivo, MEMNON.
• Core, C., Hoff, E., Rumiche, R., and Señor, M. (2013). Total and Conceptual Vocabulary in Spanish-English Bilinguals From 22 to 30 Months: Implications and Assessment, Journal of Speech, Language, and Hearing Research, 56(5), 1637-1649.
• Cousins, L. (2009). Yummy: my favorite nursery stories, Walker Books; Portuguese translation by Oliveira, J. (2010). Que delicia! As minhas histórias infantis preferidas, Editorial Caminho; Japanese translation by Haijima, K. (2010). Pakkun! Oishii Mukashi Banashi, Iwasaki Shoten.
• e-Stat, https://www.e-stat.go.jp/stat-search, (Viewed on March 12, 2020).
• Shimoida, K. (2014). Speech-Language Assessment of Culturally and Linguistically Diverse (CLD) Children in California: Practical Issues, Japanese Journal of Communication Disorders, 31(2), 112-119.
• Pearson, B.Z., Fernandes, S.C., and Oller, D.K. (1993). Lexical development in bilingual infants and toddlers: Comparison to monolingual norms, Language Learning, 43, 93-120.
• Sasaki, Y. (2014). The role of foreign-national childcare staff in multicultural symbiotic childcare, Child, Youth and Environmental Studies, 10(2), 58-65.
• Uccelli, P. & Páez, M.M. (2007). Narrative and Vocabulary Development of Bilingual Children from Kindergarten to First Grade: Developmental Changes and Associations Among English and Spanish Skills, Language, Speech, Hearing Service in School, 38(3), 225-236.