|
Asia-Pacific
Forum on Science Learning and Teaching, Volume 6, Issue 2, Article 2
(Dec., 2005) Sabri KOCAKULAH, Evrim USTUNLUOGLU and Aysel KOCAKULAH The effect of teaching in native and foreign language on students' conceptual understanding in science courses
|
Method
Procedure
First, a research design and a timetable concerning the pilot study, interviews with the teachers, achievement and conceptual understanding tests were identified. Necessary permissions were obtained in order to carry out the research.
In this study, causal-comparative research design was used. Causal-comparative design determines the cause, or reason, for existing differences in the behaviour or status of groups. This design is retrospective; that is, it starts with an effect and seeks its possible causes (Gay and Airasian, 2000). Since this study was conducted to determine the differences between students who took the science course in native and foreign language and the effect of language on conceptual understanding, this approach was decided to be used.
The research was conducted at two types of schools - one following an immersion program, English, and the other teaching academic courses in the native language, Turkish- because they accept students based on their success identified by a nation-wide standardized central exam. Thus, the two schools both had students with a similar level of achievement based on this exam. In addition to this central exam, an achievement test was administered to students to identify groups of equal achieving students at each school. The test aimed to evaluate the general academic achievement and to identify any significant differences between the two schools. In this study, Ss1 refers to the students who were taught in a foreign language, English and Ss2 refers to the students who were taught in the native language, Turkish.
Two science teachers participated in this study. One of them taught "The Energy Unit" in the native language, Turkish, while the other taught in a foreign language, English. It was not possible for researchers to find a teacher who was able to teach at both schools because of the specialized foreign language and science training required to teach the science course in English in Turkey. Thus, researchers compared the method both teachers applied in the classroom, their experiences, the curriculum, and the materials they used in the classroom. Both teachers had over 7 years teaching experience, followed the same curriculum and syllabus, and used the same materials in the classroom. They both used traditional teaching methods, teacher-centred and mostly based on questions and presentations by the teachers.
In-class observations were made by the researchers to monitor the similarity of the teaching process.
Achievement Test
The first design of the achievement test consisted of 40-multiple choice questions about the topics taught in the 8th grade science courses. These questions were chosen from standardized tests and passed through a process of refinement and validation. To do this, the questions were revised based on the reactions on the three science teachers about face validity, clarity of language and suitability for the age level concerned. Researchers did not include questions about "The Energy Unit" in the achievement test since students had not yet learned this unit.
In order to optimize the reliability and validity of the original test, the test was first given to a pilot group of 60 grade 8 students. After necessary revisions stemming from the item analyses of the pilot study, in terms of item difficulty and item discriminatory indexes, a 25-question achievement test was formed. This refined version of the test had Coefficient Alpha (or KR-20) of 0.87 and average item difficulty index of 0.65.
Before the "Energy Unit" was taught, the finalized achievement test was administered simultaneously to five 8th grade classes at the English medium school and five 8th grade classes where academic courses were taught in Turkish. Five classes were chosen from each school to ensure that the five classes at each school had the same science teacher. ANOVA (Analysis of Variance) was used for statistical analysis of the achievement test. The results indicated that there were significant differences among the groups in both schools [F=9,616; (p=,000<.05)] (see Table I). In order to identify which groups were equal to each other, LSD (Least Significant Difference) test was applied. Statistically, three groups from Ss1 and three groups from Ss2 were found to be equal. Table II shows that the means of the second (20.1579), the fourth (20.0588) and the fifth (20.8571) groups from Ss1 are close to the sixth (20.3824), seventh (20.0000) and the tenth (20.5946) groups from Ss2.
Table I. Comparison of 8th Grade Students Based on the Classes at Both Schools In Terms of Achievement
Table II. The Results of LSD Significance Test of 8th Grade Students Based on the Classes at Both Schools.
20,5946 4,016* 2,723* 3,650* 3,817* *The mean difference is significant at the .05 level.
After 3 equal groups at each school were identified statistically, a conceptual understanding test was given to the students after "the Energy Unit" was taught to both groups in the second term by two science teachers.
Participants
After the achievement test had been analysed, total 214 students coming from those 3 classes at each school participated in the study. 107 of the students were from the immersion programme secondary school. 63 were male and 44 were female students. The remaining 107 students were from the native language secondary school 58 were male and 49 were female students.
Instrument
The conceptual understanding test: During the process of structuring the instrument, concepts in the energy unit were first identified and a concept map was formed (see Figure 1). Meanwhile, literature about energy was reviewed to prepare the questions covering those concepts. It was decided to include six questions in the instrument, two of which were adopted from the studies of Brook and Driver (1984) and the rest of the questions were designed by the researchers. Additionally, experts' opinions were taken into consideration and the questions were translated from English into Turkish by a linguist.
The conceptual understanding test was tested on two focus groups, one from the school taught in Turkish and one from the school taught in English for piloting purposes. After interviews with experts in the field and students about the comprehensibility and clarity of the questions, the questions were revised and only four questions (see Appendix) were included in the test. Figure 1 shows the connection of the questions in the conceptual understanding test to the related concepts.
Questions were asked in the native language, Turkish. The reason for this was to reveal any difference between the two groups of students taught in Turkish and English and to identify whether students had assimilated "the Energy Unit" conceptually. If the aim were to evaluate how much they could understand in English, this would be measured with a reading comprehension paragraph. Taking those reasons into consideration, researchers decided to give the instrument in Turkish.
Figure 1. Concept map for energy unit and the distribution of test questions.
Data Analysis
The student responses obtained from the conceptual understanding test were coded in different categories using the same categorization process as that employed by Driver and Erickson (1983). First, the nomothetic approach was used to identify a set of scientifically acceptable response categories together with the correct response to each question developed by the researchers and experts in the field. Then, the students' answers to the open-ended questions were categorized according to scientifically acceptable or unacceptable ideas. Coding continued ideographically (Kocakulah, 1999). The response categories in the scientifically unacceptable group were later developed. Under this category, different ideas were classified into mutually exclusive sets according to the ideas related or unrelated to energy. The total percentages of students in each response category were recorded. The coding system used in the analysis of four questions is illustrated in Table III, which shows the two main categories - "scientifically acceptable arguments" and "scientifically unacceptable arguments" with their sub-categories.
Table III. Coding System Used for Students' Responses to Conceptual Understanding Test
A. Scientifically Acceptable Arguments 1. Full Argument 2. Part of Argument B. Scientifically Unacceptable Arguments 1. Response Related to Energy 2. Response Related to Other Inappropriate Ideas C. Uncodeable response D. No Response
Copyright (C) 2005 HKIEd APFSLT. Volume 6, Issue 2, Article 2 (Dec., 2005). All Rights Reserved.
