Abstract
English language education was introduced into Lao secondary schools in 1986 under the “Chintanakan Mai” (new way of thinking) policy. Since then, the demand for it has been steadily increasing. This demand, however, has not been matched by a corresponding quality improvement. Although it is widely recognized that Lao secondary school students’ achievement in English is not very impressive, there is no actual data to support this claim. In order to improve Lao secondary school students’ proficiency in English, this research aimed to ascertain the students’ level of English and to identify the factors which might impact on their achievement. For this purpose, an English test and questionnaire was administered to 1170 grade 5 students in secondary schools. This accounts for 3.1% of all grade 5 students in Laos. The results suggest that 1) English language education in Lao secondary schools should focus more on the learning of basic vocabulary, and 2) the content of the textbooks should be more basic.

Keywords: English achievement, Influential factors, Secondary education, Laos

1. Introduction
In 1986, Laos adopted the “Chintanakan Mai” (new way of thinking) policy, representing a change from a centrally planned economy and an Eastern bloc-oriented foreign policy toward a market economy and an omni-directional foreign policy. Since the implementation of this policy, an inflow of people and capital from the West has raised the importance of and the demand for English language learning. For instance, our survey found that 81.2% of Lao secondary school students “strongly agree” and another 12.0% “agree” that English is important; when English is compared with other subjects, 37.2% answered that they “work on English harder than any other subject” while 49.1% “work on English as hard as other subjects” and only 13.6% “work on English less hard than any other subject (1)”. Currently, many Lao students are anxious to be employed by foreign-affiliated companies, which usually offer salaries more than five times that paid by local companies or public agencies, and where English language skills are often required as a condition of employment.

The demand for English is increasing within the government as well, as it seeks to recruit officers with English skills in order to attract more foreign capital for economic development. As a result, English language education was introduced into the Lao secondary school curriculum in 1986. English, together with French and Russian, became one of the few elective foreign languages in the national curriculum, prescribed in 1994 by the National Research Institute for Educational Sciences (NRIES, 1994).
 
2. English language education in Lao secondary schools
Improvement in the quality of English language education has not kept up with the rapidly increasing demands for English in Lao secondary schools. There is a lack of textbooks, qualified English teachers and also a well-organized curriculum (Goh & Vonchith, 2003). As a result, it is widely believed that student achievement at secondary school level in English is unsatisfactory. However, as a large-scale study of student achievement at secondary school has never been undertaken, a conclusive statement cannot be made at this point.
In the world’s under-developed countries, such as Laos, theories of educational improvement introduced by developed countries are often accepted at face value. Such countries are often required to follow the assistance policies of the donor countries, which may not necessarily reflect their actual situation or needs. These practices impede the determination of the real causes of educational problems and one way of addressing them. In Laos, Teaching English to Speakers of Other Languages (TESOL) has already been introduced in English classes at some secondary schools. But, according to this study, the employment of TESOL, as one of teaching methods for English, appears to be too difficult for the level of English proficiency achieved by Lao secondary school students. 　　

The relevant literature regarding English education in Lao secondary schools is severely limited, especially compared to other Asian countries. If there is any, the work is not based on findings from field surveys and only describes the history or general situation of English language education in Lao secondary schools (e.g., Appleby et al. 2002; Goh & Vonchith, 2003; Sithirajvongsa & Goh, 2004).

Thus, in order to improve English educational policy in Laos, it is important to initially clarify the actual level of English proficiency of Lao students, and identify the factors which affect student achievement in English with research data.
  
3. Research method
3.1. Schedule and implementation
Both the test and the questionnaire survey were conducted in September 2003. Under the supervision of Souvannasy Bouangeune and his four assistants, the tests and the questionnaires were implemented at the sample schools. The assistants were given instructions before arriving in the field; they followed a written manual and used checklists in implementing the survey.

3.2 Sampling areas, schools and students
In this study, we administered a test at the beginning of the school year to Grade 5 students, who had been learning English for at least four years ― from Grade 1 of secondary school to Grade 4, the first year of upper secondary school. The expectation was that, after learning English for four years, students would have reached a certain level of proficiency.
The total numbers of schools in Laos consist of 847 lower-level secondary schools and 270 upper-level secondary schools, with gross enrolment rates in 2002 of 51.9% and 30.0% respectively (Ministry of Education, 2003). Among the 270 upper-level schools, 32 were independent upper secondary schools with students from Grade 4 to 6, and 238 were integrated secondary schools with students from Grade 1 to 6. Both types of schools accept students based on their scores in the provincial achievement test at the end of Grade 3, but the integrated secondary schools also allow their Grade 3 students to advance to Grade 4 based on school tests.
For reasons of accessibility, the researchers chose Vientiane Municipality and Vientiane Province from the 18 provinces in Laos as survey areas. In Vientiane Municipality, four urban districts out of nine were excluded because the socioeconomic situation and conditions in the schools were very different from other areas in Laos. Even so, the two sample areas were considered to be privileged compared to other provinces, because schools in these areas are assumed to have, in general, more quality teachers. In mountainous provinces, many secondary schools cannot provide students with English as an elective subject due to a lack of qualified teachers.

To control for student aptitude, the researchers excluded the three best schools of the 31 upper secondary schools in the five target districts of Vientiane Municipality (Vientiane Municipality Education Office, 2003) and the two best schools of the 26 schools in Vientiane Province (Vientiane Province Education Office, 2003). These schools attract high-achieving students from large catchment areas and were identified by several officials in the provincial/municipal offices of education. The researchers also excluded those schools that had changed English teachers during the last school year (Grade 4).  Schools that had tracking classes were also excluded, because it was difficult to do a random sampling of students in these schools.

As a result, twenty-five schools were selected: 15 in Vientiane Municipality and 10 in Vientiane Province. Twenty-one of the schools were integrated secondary schools and four were independent upper secondary schools. In each school, the researchers selected one Grade 5 class using random sampling. Thus, the sample size in terms of classes was 25. The total number of students was 1,170.

The twenty-five schools comprise 9.3% of all 270 upper secondary schools in Laos, and the 1,170 students make up 3.1% of all Grade 5 students (37,665 in total).

3.3. English Test
The researchers developed a test to measure student achievement and the teachers’ knowledge of English. At first, we tried to use a standardized test consisting of grammar, vocabulary and communication (Tanaka & Lauer, 2003a; Tanaka & Lauer, 2003b), with small modifications based on Lao textbooks. However, the results of a pilot study revealed that this test was too difficult for the Lao students. Consequently, the researchers were obliged to almost totally rewrite the test.

The rewritten test had three parts: grammar and sentence patterns (20 items), vocabulary (20 items), and reading comprehension (5 items). Due to limitations in terms of time and equipment, the test did not include pronunciation. The content of the test was based mainly on Textbook 1 and 2, which were expected to be completed during Grade 1 to 3.
A multiple-choice format offering four answer choices for each question was used. Grammar and sentence patterns included the usage of the object pronoun; wh-questions; prepositions; comparative and superlative adjectives; tenses (present simple; present continuous; present perfect; past simple and future with going to + infinitive verb); quantifiers, plural nouns; requests with could; yes/no questions; and the passive voice and verb agreement. Twenty items accounted for one point each. Sixteen of these items appeared in Textbook 1 (NRIES, 1996) and four items in Textbook 2 (NRIES, 1997b).
The vocabulary section consisted of 20 items corresponding to 20 basic vocabulary words. Twelve of the words were selected from amongst the 1,000 most frequently used words of the General Service Word List (West, 1953), two were selected from words 1,001 to 2,000 of the list, and another two from words 2,001 to 2,284. According to Schmitt (2000), a vocabulary of 2,000 words is considered to be necessary for successful daily communication. The four remaining words (“three”, “slim”, “wedding” and “get up”) were not taken from the list. Fifteen of the words appeared in Textbook 1 (NRIES, 1996) and three were from Textbook 2 (NRIES, 1997b). The remaining two words appeared in the earlier chapters of Textbook 3 (NRIES, 1998b).

The reading comprehension section consisted of one simple passage of text (120 words) extracted from Textbook 2 (NRIES, 1997b, p. 115). Five items accounted for two points each.

3.4. English test scores
The analysis treated English test scores as the dependent variable. Table 1 shows the descriptive statistics of the student scores on the test. The mean score was 16.4 out of a full score of 50 (32.8%). Considering the fact that the test was in a multiple-choice format with four answer choices, and that the rate of “no answer” was only 3.6%, the student scores were considered to be low.

Many Lao students have serious problems with basic vocabulary; a mean score of 7.5 out of a full score of 20 (37.5%) reflects this. It is striking that many students, after studying English for four years, still do not know the meaning of basic words like “know” and “water”. Vocabulary is the part of the test that accounted for the largest difference in student scores, with a standard deviation of 4.0.

Many Lao students also have serious problems with regard to sentence patterns/grammar and reading comprehension tasks. These problems were reflected in a mean score of 6.7 out of a full score of 20 (33.5%) and 2.2 out of a full score of 10 (22%) respectively.

Table 1: Descriptive Statistics of English Test Scores

3.5. Questionnaire and variables
Educational quantitative researchers look at different factors related to student achievement, depending on their own interests. To reveal the factors related to student achievement from broader viewpoints, combining different kinds of research approaches, rather than focusing on a specific factor, such as parental attitude toward education (a family factor) or learner-centered teaching methods (an educational factor), would be most effective. With this approach, however, the research would fail to provide a detailed description of how a specific factor works in a certain situation. So, to reveal how a specific factor works in a certain situation, a detailed description would be more useful; however, this approach may not provide an overall image of how each factor interrelates with student achievement from different kinds of factors’ viewpoint. To make up for this, the results obtained by each kind of research method can be aggregated.

In Laos, where there is a limited amount of extant educational research on English language education, and where the students’ main purpose for studying is to obtain high grades in order to gain entrance into institutes of higher education, a broad approach to English achievement which examines the relationship between various factors directly related to student achievement would be most instructive. Therefore, this research examines the related factors by combining three different kinds of factors: family factors (family background), motivation factors (motivation for English study) and educational factors (educational inputs). Previous research on the cause and effect relationship with regard to student achievement has shown that these kinds of factors are important in assessing student achievement (2).

Questionnaires were distributed to students who took the English test. The questionnaire consisted of 20 questions (Table 2). Nine of the questions related to family background; from the answers, four variables were created. Four of the questions related to motivation for English study; from the answers, four variables were created. The seven remaining questions related to educational inputs; from the answers, five variables were created. The variables created are often cited as basic variables in research on family background, motivation for English study and educational production. An evaluation scale for all 13 variables except for “average score in a class”, was employed for the answered questions (3).

Table 2: Descriptions, Means, and Standard Deviations for Variables

4. Data analysis
As mentioned above, based on the answers to the 20 questions in the questionnaire, a total of 16 variables were created. The average score in each class was then added as an independent variable, and English test scores were added as a dependent variable. The data obtained was analyzed statistically, using correlations and a multiple linear regression equation.

4.1. Correlations between the variables
The correlations between each variable, which are deemed to be important for student academic achievement, were determined by means of bivariate correlations. Although bivariate correlations do not necessarily indicate a relationship of cause and effect, they are expected to give an overall indication of which variables are related with which. Table 3 shows that many of the variables, especially the relationship between English test scores and other variables, were positively correlated with each other at a significance level of 5%. It can thus be inferred that these variables ― family background, motivation for English study and educational inputs ― have some influence on the English test scores achieved (4).

Table 3: Correlations between the Variables  

Table 3: Correlations between the Variables (Continued)

4.2.1 Standardized regression coefficients from multiple regressions of English test scores on variables
To determine which variables strongly affected English test scores, a multiple linear regression equation was employed. English test scores were regressed on variables (Table 4). Six of the seven significant variables related to educational inputs and it appears that educational inputs have a stronger effect on students’ English test scores than the other two kinds of variables. The average score in a class, especially, seems to have the strongest effect on English test scores (5).

Extra English classes or private tuition is also likely to be one factor with a strong effect on English test scores. Extra classes are usually held after school, for one period (one hour) a day, five days a week during the semester. Six months of extra classes are expected to provide about 85 periods of teaching. The English teachers at the schools usually teach the extra class. Students pay a fee for the extra class, which goes to the teacher. In most cases, all students, regardless of their grade level or their previous experience in attending extra classes, are taught together in the same class. Students in Grade 5 have studied English at school for more than 396 periods, as they have had three periods a week for 33 weeks a year over four years. It is difficult to believe that by adding 85 periods to the 396 periods, student scores could improve dramatically. The results suggest that the effect of extra classes is not only the result of additional teaching time, but also of the quality of the teaching. It is widely recognized that in the extra classes, many teachers use imported textbooks that are better-structured and more systematic in teaching sentence patterns/grammar.

Table 4: Standardized Regression Coefficients from Multiple Regressions of Total English Test Scores     

Possession of English textbooks also indicates a significant effect on English test scores. Of course, it is natural that reading English books improves English ability, especially reading ability. However, judging by the test scores of the survey, it seems that even a simple English book would be too difficult for these students. Therefore, more importantly, possession of English books might mean that families with English books have a more positive attitude toward English, and that some other family members are also familiar with English.

On one hand, some educational inputs are very likely to have an effect on English test scores but, on the other hand, some others are not. It is important to note that problems exist in the educational inputs with little or no effect on English test scores; why aren't these inputs working effectively? Of all these factors, in this research, possession of textbooks gives a result particularly contrary to previous research, and is worthy of further study in a Lao context.

Many previous studies have revealed that possession of textbooks is an important input for improving student achievement in developing countries (e.g., Fuller & Clarke, 1994; Harber & Davies, 1997; Scheerens, 1999). However, the content of the English textbooks used in Laos seems too difficult for Grade 5 secondary school students. Figure 1 shows the results from the vocabulary section of the English test given to Grade 5 students in the study. The test results demonstrate that many students are not familiar with basic vocabulary items; for instance, 66.3 % of students did not know the meaning of “meat”; 53.3% did not know the meaning of “bicycle”; 51.5% of students did not know the meaning of “strong”.

A comparison of the test results and the content of the textbook reveal a mismatch between the students’ English level and the content of the textbook. For example, Picture 1 shows pages 2 and 3 of Textbook 3, used by Grade 5 students in Laos (6). In spite of the fact that many Grade 5 students are not familiar with some basic vocabulary items, the readability of the textbook expects them to understand long sentences with more difficult vocabulary items. Furthermore, most of instruction and explanation are written in English (7).

Figure 1: Rate of Correct Answers in Basic Vocabulary*

4.3.1 Standardized regression coefficients from multiple regressions of English test scores of each section on variables
Next, English test scores were classified into the three sections - vocabulary scores, grammar scores and reading scores to compare the differences in the effect of the same variables on each section (Table 5). Table 5 is the result of a multiple linear regression equation (Enter method) in which vocabulary scores, grammar scores and reading scores became dependent variables. Comparing the results of vocabulary scores as a dependent variable with the results of grammar scores and reading scores as dependent variables, for vocabulary scores, the family economic indicator is not significant, but possession of an English-Lao dictionary is significant, while possession of English-Lao dictionary for grammar scores and for reading scores (Table 5: Step 1) are not significant. This result partly shows that, unlike the results for textbooks, possession of an English-Lao dictionary contributes to mastering vocabulary. In general, textbooks need to target a specific level of learner, while many English dictionaries, including advanced dictionaries, can be used by different levels of learners (8). For this reason, a Lao-English dictionary might work effectively. 

Stanovich (1991) pointed out that poor reading comprehension could occur even with a large vocabulary in terms of size, but a poor vocabulary never leads to good reading comprehension. Reading scores (Table 5: Step 2) are the result of a multiple linear regression equation in which vocabulary scores and grammar scores are also added to the independent variables (Table 5: Step1). Three variables emerge at the significance level of 1%. Of the three, vocabulary scores show the highest standardized coefficients. The strongest effect on reading scores is shown by vocabulary scores as an independent variable. Considering the poor vocabulary of Lao students, the necessity of vocabulary learning should be emphasized for reading comprehension.

While possession of English books is significant for vocabulary scores and grammar scores, it is not significant for reading scores (both Steps 1 and 2). This result confirms the aforementioned inference that possession of English books does not lead to an improvement in students' reading ability. There may be another reason that possession of English books is significant only in vocabulary scores and grammar scores.

Table 5: Standardized Regression Coefficients from Multiple Regressions of Vocabulary Scores, Grammar Scores and Reading Scores on Each Variable

5. Summary
This study has revealed several thought-provoking findings, with ramifications for English language education in Laos. First, Lao secondary students have a problem with basic vocabulary. Moreover, as vocabulary skills seem to have a significant effect on reading comprehension, basic vocabulary learning should be emphasized before reading comprehension is practiced. In order to do this, an English-Lao dictionary could be an important tool. In addition, the content of the textbook should be revised to focus more on basic learning, and it would be better to use Lao, rather than English, for instructions and explanations. As grammar scores are also low and are likely to affect reading comprehension, students should also focus more on basic grammar before tackling reading comprehension. 

Endnotes:
(1) Because the percentages are rounded up to the second decimal place, total percentages do not equal 100%.
(2) For example, regarding family factors, refer to Kellaghan (1994) or Buchmann & Hannum (2001). Regarding motivational factors, refer to Joseph et al. (1997) or Lens (1994). Regarding educational factors, refer to Fuller & Clarke (1994) or Scheerens (1999).
(3) In the research, the variable “average score in a class” is considered to be an integrated indicator of the effects of a variety of school factors such as class facilities, class climate, leadership of the principal, teaching quality, and so on. These were not examined in the research. Further detailed investigation is needed in order to clarify which of the school factors contributed to the results.
(4) Results of the t-test show that gender differences in English test scores were not observed at a significance level of 5%. The average scores of male and female students were 16.23 and 16.73 respectively.
(5) This might not be surprising, since the average score in a class comprises an individual student's English test score. Nevertheless, as the relationship only accounts for one of approximately 46.8 students (the average number of students per class in the sampling schools), the effect of an individual student's English test score on the whole relationship is not large.
(6) The total number of pages in the textbook is 162. Secondary students are expected to finish the entire textbook by the end of the first semester of Grade 5. On the whole, with each increasing page number the content of the textbook becomes more difficult.
(7) In fact, it is doubtful that the teachers themselves fully understand the textbooks.  Twenty-four English teachers, who taught the students who participated in the study, were asked to take the same test as the students. They only answered 82% of the questions correctly. For the reading comprehension section, they only answered 64% of the questions correctly. It can be thus inferred that the lack of ability of teachers poses limitations on their ability to successfully teach students.
(8) For instance, students learning English do not usually change English dictionaries every year, although they change English textbooks for higher level ones every year.

References
Appleby, R., Copley, K., Sithirajvongsa, S. & Pennycook, A. (2002). Language in development constrained: Three contexts. TESOL Quarterly, 36(3), 323-346.

Buchmann, C. & Hannum, E. (2001). Education and stratification in developed and developing countries: A review of theory and research. Annual Review of Sociology, 27(1), 77-102.

Fuller, B. & Clarke, P. (1994). Raising school effects while ignoring culture? Local conditions and the influence of classroom tools, rules, and pedagogy. Review of Educational Research, 64(1), 119-157.

Goh, E. & Vonchith, T. (2003). English language teaching in Laos. In H. W. Kam & R. Y. L. Wong (Eds.), English language teaching in East Asia today: Changing policies and practices (pp. 257-268). Singapore: Times Academic Press.

Harber, C. & Davies, L. (1997). School management and effectiveness in developing countries. London: Cassell.

Joseph, O. J., Ruth, T. J. & Christianah, A. U. (1997). Effects of achievement motivation and study habits on Nigerian secondary school students’ academic performance. Journal of Psychology, 131(5), 523-529.

Kellaghan, T. (1994). Family and schooling. In The international encyclopedia of education (2nd Ed., Vols. 4, pp. 2250-2258). Oxford: Pergamon Press.

Lens, W. (1994). Motivation and learning. In The international encyclopedia of education (2nd Ed., Vols. 7, pp. 3936-3945). Oxford: Pergamon Press.

Ministry of Education. (2003). Bothsalub Kanjathtang Patibath Phanphatthana Kanhsuksa 2002-2003 Lae Phanphatthana Kanhsuksa 2003-2004 [Summary of the educational implementation plan for the academic year 2002-2003 and educational plan for the academic year 2003-2004]. Vientiane.

National Research Institute for Educational Sciences, Ministry of Education. (1996). English for Lao secondary schools 1, Vientiane: Printing Press of the Ministry of Education.

National Research Institute for Educational Sciences, Ministry of Education. (1997a). Laksuth Phasa Angkith Sanmatthayomton [English curriculum for lower secondary schools], Vientiane: Ministry of Education.

National Research Institute for Educational Sciences, Ministry of Education. (1997b). English for Lao secondary schools 2, Vientiane: Printing Press of the Ministry of Education.

National Research Institute for Educational Sciences, Ministry of Education. (1998a). Laksuth Phasa Angkith Sanmatthayon Tonpay [English curriculum for upper secondary schools]. Vientiane: Printing Press of the Ministry of Education.

National Research Institute for Educational Sciences, Ministry of Education. (1998b). English for Lao Secondary Schools, Vientiane: Printing Press of the Ministry of Education.

Reena, S. & Peter, B. (2001). Reading in a second language: Factors associated with progress in young children, Educational Psychology, 21(2), 189-202.

Scheerens, J. (1999). School effectiveness in developed and developing countries; A review of the research evidence. World Bank Paper, Retrieved on February 4, 2006 from
http://www1.worldbank.org/education/est/resources
/topic%20papers/jaap699.doc

Schmitt, N. (2000). Vocabulary in language teaching. Cambridge: Cambridge University  Press.

Sithirajvongsa, S. & Goh, E. (2004). Lao PDR. In H. W. Kam & R. Y. L. Wong (Eds.), Language policies and language education: The impact in East Asian countries in the next decade (2nd Ed., pp.196-206). Singapore: Eastern Universities Press.

Stanovich, K. (1991). Word recognition: Changing perspectives. In R. Barr, M. L. Kamil, P. Mosenthal & P. D. Pearson (Eds.), Handbook of reading research (pp. 418-452). New York: Longman.

Tanaka, M. & Lauer, J. (2003a). English qualification test for Japanese university students. Hiroshima: Hiroshima University.

Tanaka, M. & Lauer, J. (2003b). Vocabulary levels test for Japanese learners of English. Hiroshima: Hiroshima University.

Vientiane Municipality Education Office. (2003). Bothsalub Kanjathtang Patibath Phanphatthana Kanhsuksa 2002-2003 Lae Phanphatthana Kanhsuksa 2003-2004 [Summary of Educational Implementation Plan for the Academic Year 2002-2003 and Educational Plan for the Academic Year 2003-2004]. Vientiane.

Vientiane Province Education Office. (2003). Bothsalub Kanjathtang Patibath Phanphatthana Kanhsuksa 2002-2003 Lae Phanphatthana Kanhsuksa 2003-2004 [Summary of Educational Implementation Plan for the Academic Year 2002-2003 and Educational Plan for the Academic Year 2003-2004]. Vientiane Province.

West, M. (1953). A general service list of English words. London: Longman, Retrieved on February 4, 2006 from http:jbauman.com/gsl.html