Text Comprehension Strategies of Children
in an Indonesian Bilingual Program

The ACIE Newsletter, May 2002, Vol. 5, No. 3

By Michèle de Courcy, Institute for Education, LaTrobe University, Bendigo, Victoria



A particular focus of recent language education policies in Australia has been the teaching of Asian languages, and Indonesian has been selected as one of the priority languages.

During 1998-9 I conducted a pilot study at a primary school in a country town in northern Victoria that is running a content-based Indonesian program - the only such program in an Australian primary school. As the children are taught in Indonesian for only 35% of the school timetable, the program is classified as ”bilingual,"rather than ”immersion."There are six classes of students involved in the program - two at Year 1/2, two at Year 3/4 and two at Year 5/6. An unusual feature of this program was that all these classes were started in the bilingual program at the same time - thus, at the time when the study was undertaken, even the year 5/6 group had only been learning Indonesian in the bilingual program for one to two years. The main content area delivered through the second language is Mathematics, with some sections of other Key Learning Areas (KLAs) also being offered, notably Dance and Studies of Society and the Environment.

The project aimed to investigate the text-comprehension strategies, especially the role of the first language (L1 - English) and the second language (L2 - Indonesian) of children in the Year 5/6 classes at the school.


The main part of this pilot project involved observations of the children's reading and problem-solving behaviors in the Mathematics classroom. Children were observed using the natural thinking aloud processes they employ to work on problem solving strategies, e.g. counting on fingers or talking to themselves as they solve the problem, as suggested by Cohen (1998). Extensive field notes and audio taping were used in the classrooms. Meetings were also held with the teachers during the period of the study. An attitudinal questionnaire asked children about their language background, learning preferences, and self evaluation in Indonesian.


The only literacy tasks the Year 5s were confident of being able to do were read things on the board and instructions for Maths homework. The Year 6s, who were in the second year of immersion, were much more confident, and felt they could do nearly all of the tasks suggested on the questionnaire. The year 6 children's ranking of these tasks from most to least confident was:

  • Read and understand a letter or note written in simple Indonesian
  • Read and understand what my teacher writes on the board in Maths class
  • Read and understand a story in simplified Indonesian
  • Read the instructions for doing my Maths homework in Indonesian
  • Understand a newspaper headline written in Indonesian
  • Understand a sentence I haven't seen before in a book written in Indonesian about Australia or Australian animals
  • Read and understand a factual piece written in Indonesian
  • Read an Indonesian story book written for a person of my age, without using a dictionary

The questionnaire results revealed the children's reluctance to use Indonesian outside the compulsory context of answering a teacher's questions, but also a lack of embarrassment about using Indonesian in this context. The children reported that, outside the classroom, even when addressed in Indonesian, they reply in English, and this was the pattern observed by the researcher as well.

In relation to strategies, in particular, regarding the use of first and second languages, the students were asked two questions which related to mental arithmetic. Thirty-eight children reported that they used English to count when they are alone, and eleven selected Indonesian as their first choice. Note that these children initially learnt to count in English. For the more difficult operation of adding up numbers, forty-seven of the forty-nine children surveyed reported using English to add up numbers, with only fifteen putting Indonesian as their second choice. Only two children reported using Indonesian before English.


In the classroom, the children were observed working together in small groups to solve problems, verbalizing their thinking processes, and also using drawings to aid them in the task. The children frequently turned around to consult charts on the back wall which show the Indonesian words for basic operations.

To solve word problems, the students employed the following strategies:

  1. Re-reading the problem out loud.
  2. Use of the first language to understand the problem in the second language - this is where the children use the many loan words found in Indonesian to help their understanding.
  3. Use of code mixing of the first language and the second language. They all do this to some extent.
  4. Translation of certain key words as a means of comprehension.
  5. Translation of whole phrases or sentences
  6. Use of a partner.
  7. And most frequently - ignoring the words, looking at the numbers, and deciding on the basis of the numbers alone, which mathematical operation to use.

In the class selected for illustration, the children were solving word problems on a worksheet related to the current theme of ”space.’ I noted that they formed natural cooperative groupings, and I chose to observe and record two boys working together. Each question was first read aloud by one of the boys, after which they employed various strategies to solve the question.

In question 3 below, the boys used strategies i, iv, vi, and vii. (Note that the students were not provided with the translations given here to aid the reader!)

Question 3. Sembilan astronot menghabiskan total waktu 1337 jam didalam simluator. Berapa jam masing-masing astronot itu tinggal didalam simluator? __________ jam.
(Nine astronauts spent a total of 1337 hrs in a simulator. How many hours did each astronaut stay in the simulator?)

We see first the use of a partner to make sense of unfamiliar key words in the problem:

Extract 1
C1 it's got 'jam'- what's jam?
C2 what's 'itu'?
C1 it's 1337 divided by 9 maybe
C2 masing-masing - I knew it (but can't now)

The boys focussed initially on words they thought were key words to problem. When they couldn't understand them, they decided, on the basis of their experience with numbers, that they needed to divide 1337 by 9 (which was, luckily, the correct operation).

The second strategies illustrated here are ii and v. Quite often the children would write out a translation of the word problem on their scratch pads to help them work out the answers. In question 4, as illustrated below, one of the boys had written in English over the Indonesian words, ”How many years after discovered ...and ... .’ The word ”Neptunus’ was close enough to the English word ”Neptune’ for the student not to bother translating it. This problem was solved correctly.

Question 4. Uranus ditemukan pada tahun 1781. Neptunus ditemukan pada tahun 1846. Pluto ditemukan pada tahun 1930. Berapa tahun jarak antara ditemukan Uranus dan Neptunus? __________ tahun.

How many years after discovered ...and ...

Berapa tahun jarak antara ditemukan Neptunus dan Pluto? _______ tahun.
(Uranus was discovered in 1781. Neptune was discovered in 1846. Pluto was discovered in 1930. How many years after Uranus was discovered was Neptune found? ____ years. How many years after Neptune was discovered was Pluto found? ____ years.)

The final example is one which illustrates a typical error made by the children in bilingual classrooms - performing a one stage operation for a two-stage problem (Sherril, 1983). This is caused in this case by application of strategies iv and vii.

Question 5. Pesawat A panjangnya 36.25 m. Pesawat B, 4.09 m lebih panjang dari Pesawat A. Pesawat C, 5.78 m lebih panjang dari pesawat B. Berapa total panjang keitga pesawat itu? ___________ meter
(Airplane A measures 36.24 m. Airplane B is 4.09 m longer than Airplane A. Airplane C is 5.78 m longer than Airplane B. What is the total length of all three airplanes?)

The task was to work out the total length of all three airplanes. This was a two stage operation, involving, first, working out how long each airplane was, then adding up the different lengths obtained.

With this question, the boys I was observing had first guessed that 'panjangnya' was how low the airplane flew. They guessed this because they thought that 36.25 m was rather low if 'm' was meters, therefore it must be miles (!) This is an example of how focussing on a key word and translating it incorrectly ('panjang' is length and 'lebih panjang' is longer) can lead to a totally wrong operation being used and a wrong solution resulting. During correction time, the teacher needed to draw a diagram on the board to explain question 5, as many children had been confused.


There has, of course, been a great deal of research into problem solving behaviors of children learning mathematics in their first language. For example, Newman (1977, 1979) conducted think aloud protocols with sixth grade first and second language speakers of English. She discovered a hierarchy of performance strategies which a child needs to correctly solve a Maths word problem, ranging from word and symbol recognition to being able to work out that the question itself was faulty. In this study, using Newman's terms, the children have shown problems with word recognition, comprehension related to specific terminology, general meaning comprehension and transformation.

Large amounts of translation are still being used as a receptive strategy by these students, who are still in the early stages of second language learning. This is similar to my finding with French late immersion students in 1993. Then, students reported (and demonstrated) that translation as a receptive strategy was used for about the first year of their program, being abandoned in favor of focusing on key words and still later reading for gist. In the examples shown here we see some examples of translation of perceived ?key words' and some of whole sentences.

It is important to note that the ”translation’ being used by the children is the process-type translation described by Kern (1994), which is a mental operation used to aid comprehension. This is different from the traditional type of translation exercises, whose primary purpose is not that of comprehension.

The simple operation of counting is usually carried out in Indonesian in the classroom, and sometimes when the child is alone, but more complicated operations are carried out in the first language. These children, who already have their basic mathematical concepts firmly established in their first language, are showing themselves to be efficient solvers of mathematical word problems.

From a reading perspective, the most disappointing aspect of these results is that, as hypothesized, the children are not yet really reading in Indonesian. They are either translating whole sentences or key words into English to help their understanding of the problem or ignoring the words and solving the problem based on the numbers alone. In the terms used in my earlier research (de Courcy & Burston, 2002), these children are still in the ”translation as a receptive strategy’ and ”key words’ stages. It will take more in depth and longitudinal studies in order to explore the strategies used by students at more advanced stages of the second language development.

Much more work needs to be done in this area - that is, what immersion students actually do in order to process the written text with which they are presented in immersion classrooms - before we can really understand the processes of meaning making in immersion. It is hoped that a better understanding of these processes will benefit teachers and researchers alike.



Burston, M. (1999). Mathématiques en immersion partielle: Comment les enfants s'y prennent-ils pour résoudre un problème? Le Journal de l'Immersion Journal, 22, (1), 37-41.

Cloer, T. (1981). Factors affecting comprehension of math word problems: A review of the research. American Reading Forum, Sarasota, FL, ERIC.

Cohen, A. (1995). The role of language of thought in foreign language learning. Working Papers in Foreign Language Learning, 11, 2.

Cohen, A. (1998). Strategies in learning and using second languages. London: Longman.

de Courcy, M. (1997). So, translation is indeed a receptive strategy? Response to Arcidiacono. AALIT Journal, 5, 4-7.

de Courcy, M. (1998). Bilingual and Immersion SIG. Languages Victoria, 2, (2), 16.

de Courcy, M. C., & Burston, M. (2000). Learning Mathematics through French in Australia. Language and Education, 4, (2), 75-95.

Department of Employment, Education and Training. (1991). Australia's Language: The Australian Language and Literacy Policy. Canberra: Australian Government Publishing Service.

Djité, P. (1994). From language policy to language planning: An overview of languages other than English in Australian education. Canberra: NLLIA.

Kern, R. (1994). The role of mental translation in second language reading. Studies in Second Language Acquisition, 16, 441-461.

Kirkpatrick, A. (1995). Learning Asian languages in Australia: Which languages and when? BABEL, 30, (1), 4-11, 26-29

MACLOTE (1994). LOTE report to the Minister for Education. Melbourne: Directorate of Schools Education.

Read, J. (1999). Immersion Indonesian at Rowville Secondary College. Babel: Journal of the Australian Federation of Modern Language Teachers Associations, 34, (2), 4-9, 37.

Read, J. E. (1993). Case study of an Indonesian language immersion program at Rowville Secondary College. Applied Linguistics. Melbourne: University of Melbourne.

Sherrill, J. (1983). Solving textbook mathematical word problems. Alberta Journal of Educational Research, 29, (2), 140-52.

Worsley, P. (1994). Indonesian/Malay. Melbourne: NLLIA.

CARLA Mailing List Signup Contact CARLA CARLA Events Donate to CARLA CARLA on Facebook CARLA on YouTube Twitter
Center for Advanced Research on Language Acquisition (CARLA) • 140 University International Center • 331 - 17th Ave SE • Minneapolis, MN 55414