Language and Modeling Word Problems in Mathematics Among Bilinguals

Allan B. I. Bernardo (2005)

The author investigates whether the language proficiency of bilingual students in the Philippines affects how they build problem structure in word problems. He conducted a big survey which asked 111 participating 4^th-grade students to 1) complete the several given word problems that a question sentence was removed intentionally and 2) solve the question they made up. Almost half of the participants are speaking English in their home whereas others are speaking Filipino in their home.
  One of the supported ideas of solving word problems is that firstly learners grasp textual contexts form the given problems, and they build implicit problem structure to clarify what information is known/unknown, to compare given numbers, and to find relationships among the data. Lastly, students create and execute their solution depending on the above procedures. His research shows that a language factor does not relate to the modeling problem structure (more mathematically abstract part).



All of the works in mathematics classes in Filipino schools are in English, and the author classifies the participating students into two types by home language. However, I stopped the definition of bilingual though the author does not mention it in this paper. I suppose their home language does not always match their mother tongue. For instance, some study shows that, especially for single-parent families, immigrant children talk with their parent in the first language for the parent, but they cannot communicate well because children’s first language is Japanese. I acknowledge the situations of Philippines and Japan are completely different. However, the word “bilingual” is ambiguous nowadays.
In his experiment, the students might try to generate the own questions to complete given word problems. However, those created questions should be very linguistically confined and rely on the students’ language proficiency. Because the students model problem structure based on those linguistically limited problems, it can be said there are no linguistic aspects in more mathematically abstract part. For example, can the students whose first language is Filipino solve the question made by the students whose first language is English? It is too early for me to conclude linguistic factors do not affect their modeling problem structure.