THE RECEPTIVE VOCABULARY OF ENGLISH FOREIGN LANGUAGE YOUNG LEARNERS ROSA

This paper responds to the need for research on vocabulary knowledge in foreign language education. First, we investigate the receptive vocabulary knowledge of students learning English in Spanish primary education by using the 1,000 word test and the 2,000 frequency band of The Vocabulary Levels Test (VLT). Second, we study differences between the sexes by comparing their scores. Third, we evaluate whether students’ scores correlate with their scores on a cloze test. As a result, we show that their English receptive vocabulary size falls within the 1,000 word level. Finally, we demonstrate the existence of a positive correlation between the two frequency bands and a cloze test.


INTRODUCTION
Research on vocabulary acquisition in second or foreign languages (L2) is characterized by a great deal of fragmentation as well as inconclusive results.The teacher or researcher who reads articles and books with the hope of finding answers to questions concerning vocabulary acquisition and development (such as how vocabulary develops throughout school years or what effect contextual and individual differences have on vocabulary acquisition) is often left with more doubts than certainties.However, within this apparently perplexing picture, a pattern emerges that points to the importance of vocabulary knowledge in L2 learning and to its educational and social implications.
Knowledge of the number of words known by L2 learners is crucial in any learning context but of paramount importance when such learning takes place in primary and secondary education.In these contexts, learning, as measured by tests, is going to be reflected in school grades, and as a result, is going to have an impact on students' lives.At the beginning of a new school year, teachers need to know how many words students know receptively and productively, in order to be in a position to assess students' vocabulary gains at the end of the course and diagnose possible gaps.Teachers also need to estimate their students' vocabulary size to set language levels in each course, to programme language activities and to carry out motivated selections of materials.Such knowledge is also important for test and textbook designers as well as for vocabulary acquisition researchers: for the former because they are better informed to create materials and tests suitable for different levels and educational needs, and for the latter, because empirical data from different groups of subjects can provide a baseline for comparison and help to identify patterns of vocabulary acquisition and development.
The present study is an attempt to further our understanding of young L2 learners' vocabulary knowledge by (a) surveying the receptive vocabulary size of students who are learning English as a compulsory subject in Spanish primary education, (b) investigating individual differences by means of the comparison of the scores of males and females on a vocabulary size test and (c) assessing whether their scores on a vocabulary size test correlate with their scores on a cloze test. 2

BACKGROUND
Leading scholars in vocabulary research (Nation 1990;Meara 1996;Laufer 1989Laufer , 1998;;Read 1988) believe that the number of words known is one of the key factors in L2 learning, particularly in the first stages of L2 learning where students probably have only small lexicons.Unfortunately, as Read notes, finding out how many words L2 learners know is not a straightforward issue, because when estimating learners' vocabulary size, researchers encounter conceptual and methodological problems (Read 1988).These problems have been addressed in a number of studies, such as for instance: on defining what a word is (Bauer and Nation 1993), what it means to know a word (Nation 1990(Nation , 2001;;Meara 1996), what is the minimum vocabulary size to follow academic programmnes in English as a medium of instruction (Sutarsyah, Nation and Kennedy 1994), and what is the minimum needed to understand English texts (Nation 1990;Laufer 1992Laufer , 1997;;Ward 1999).Another important issue is the selection of the test used to measure vocabulary knowledge; in this regard, a number of studies have focused on the design of receptive vocabulary size tests, among which the VLT (Nation 1983(Nation , 1990) and the Yes/No vocabulary test (Meara and Buxton 1987;Meara and Jones 1990) have had a considerable impact on vocabulary research.
The VLT (Nation 1983(Nation , 1990) ) was devised with a pedagogical aim in mind, for teachers to diagnose English learners' receptive vocabulary gaps.It contains words sampled from the 2000, 3000, 5000, The Academic Word List, and the 10000 most frequent words in English.In each of the three sections that make up the test in the five frequency levels, the testee is asked to match three definitions to six words.The main assumptions underlying the test are that the most frequent words in a language will be the first to be learned, and that vocabulary growth will take place in scalable order: that is, knowledge of words in a particular band implies knowledge of words in all lower bands, but not of those in any higher band.To put it in another way, testees' knowledge of uncommon words implies knowledge of the most frequent words but not the other way round.The VLT has been used for different purposes in a number of studies (Laufer 1997(Laufer , 1998;;Schmitt and Meara 1997;Cobb 1999Cobb , 2001)).Research has also been devoted to the validation of this test (Read, 1988), the assessment of its adequacy for secondary school learners of English as an additional language (Cameron 2002), and the elaboration of new test versions (Schmitt 1993;Beglar and Hunt 1999) together with their subsequent validation (Beglar and Hunt 1999;Schmitt, Schmitt and Clapham 2001).
As to the Yes/No vocabulary test, it was based on the previous work of Meara (1992), and Meara and Buxton (1987) as an alternative to multiple choice in vocabulary testing, and gave rise to Eurocentres Vocabulary Size Test (Meara and Jones 1990).The test format is a checklist in which test takers are presented with a list of words and asked to check whether they know each of the words within the list.To avoid false scores due to testees' overestimating their word knoweldge, a number of imaginary words are included.As in the case of the VLT, word selection and word knowledge are based on graded frequency lists.
Both tests have advantages and drawbacks but, on the whole, they have proved to be valid and reliable receptive vocabulary tests.The key issue here is whether one test is more suitable than the other to assess young learners' receptive vocabulary.In this sense what research tells us is that low level learners seem to have problems with non-words (Read 1997).According to Cameron (2002) -who studied the practicability of the Yes/No test and the VLT to investigate the vocabulary size of UK secondary school learners of English as an Additional Language -the latter is more useful for 13 to 15 year-olds because the non-words included in the Yes/No test lead learners to be confused regarding their recognition of words.On the other hand, research carried out with the VLT has given evidence of its validity when used with secondary school students in different contexts.Laufer (1998) used the test to investigate vocabulary gains in the receptive vocabulary knowledge of Israeli comprehensive high school learners of English as a foreign language.Beglar and Hunt (1999) validated four versions of the 2,000 frequency level and the University word Level with Japanese high school students.Both studies provided useful data with respect to the validity of the tests for the assessment of secondary students.
The considerable amount of time and energy devoted to the construction and validation of the most economic, valid and reliable test is perhaps one of the reasons for the dearth of studies on L2 learners' vocabulary sizes.Notwithstanding, several related research lines are found that focus on: i) the comparison between native speakers' and L2 learners' vocabulary sizes (Jamieson 1976;Izawa 1993;Cameron 2002); ii) the relation between receptive and productive vocabulary knowledge (Laufer 1998;Fan 2000) and its correlation with L2 proficiency (Fan 2000); iii) receptive and productive gains over one year of study (Laufer 1998); iv) receptive vocabulary increase throughout a study abroad programme (Milton and Meara 1995); v) estimates of vocabulary size of L2 learners (Quinn 1968;Takala 1984;Nurweni and Read 1999;Cobb and Horst 1999;Cameron 2002;Pérez 2004;López-Mezquita 2005).This line of research is extremely related to the goals of the present study.Therefore in the remainder of the section we will deal first with the characteristics and main results of these studies, then we will review research on the relationship between receptive vocabulary knowledge and the sex variable.
Studies on estimates of L2 learners' vocabulary knowledge are difficult to compare due to differences concerning subjects, the learning contexts, and the tests used for estimating vocabulary size.Whereas in Quinn (1968), Nurweni and Read (1999), Cobb and Horst (1999), Pérez (2004), the subjects are university students, in Takala (1984), Cameron (2002) and López-Mezquita (2005) they are secondary school students; however, in the latter study, two groups of university students are also investigated.In these studies, words are drawn from different sources and different test formats are used as can been seen in Table 1.
The Vocabulary Levels Laufer (1998aLaufer ( , 1998b)); Matching definitions Test (Nation 1990) Cobb and Horst (1999); to words.(words selected from Cameron (2002).Thorndike and Lorge (1944); Kucera and Francis (1967)  Surprisingly, the results obtained coincide in showing a rather low vocabulary knowledge on the part of the English learners investigated.Results speak of 1,000 words (Quinn 1968), about 1,200 words (Nurweni and Read 1999), 1,500 words (Takala 1985), the 2,000 most basic word families of English (Cobb and Horst 1999), and gaps and problems in the comprehension of the most frequent words in English (Cameron 2002).Within the context of Spanish secondary education, López-Mezquita (2005) reports an average of 941 words in 4º ESO (4th form), 1,582 words in 1º Bachillerato (5th form), and 1,855 in 2º Bachillerato (6th form).She also reports 3,174 words for first year university students of English Philology and English Translation studies.The figures reported in vocabulary size studies are low if we bear in mind that they have been produced after six or seven years of extensive study of English in high-school, and even, as in the case of Cameron's study, after 10 years of education through English.
Sex as a variable in individual differences has received little attention in L2 vocabulary research.The few studies conducted with primary and secondary school learners have shown that compared to male students, female students make use of a greater number and a wider range of vocabulary strategies (Jiménez 2003), and commit fewer lexical errors (Agustín 2005;Agustín, Fernández and Moreno 2005).Research has also provided evidence of differences in vocabulary strategy use (Jiménez 2003), choices of word topics related to social issues (Jiménez 1997), productive vocabulary in written compositions (Jiménez 1992;Jiménez andOjeda 2007, 2008) and productive vocabulary in Lex30 by Meara and Fitzpatrick (2000) (Jiménez and Moreno 2004).One of our research goals is to determine whether these trends will also appear in receptive vocabulary size studies.Unfortunately, in most of the vocabulary studies conducted so far no information is provided regarding the distribution of the informants according to the sex variable.The only data we have found on sex differences in receptive vocabulary knowledge comes from broader studies on L2, aimed at investigating the acquisition of different language skills.Hurlburst (1954) reported differences in favour of boys reflected in the mean scores achieved in word recognition and recalling tasks by males and females.Likewise, Edelenbos and Vinjé (2000) found that boys in the 8th grade of Dutch primary education outperformed girls in English word knowledge.
On reviewing research on vocabulary size in an L2, the following conclusions can be drawn: a) most research has been carried out with university students; b) few studies have been done with learners of English as a foreign language in primary education; c) there is a gap concerning studies that focus on the relationship between receptive vocabulary knowledge and individual differences such as the sex variable.To our knowledge, no empirical research has been conducted on the vocabulary size of English foreign language young learners.The present study is a preliminary attempt to fill this gap by providing data on the receptive vocabulary estimate of a large sample of 10-year-old Spanish students who are learning English in the 4th year of primary education.Our sample is highly homogeneous regarding L1s, age, proficiency level and the social profile of the areas where the schools are located.

RESEARCH QUESTIONS
a. What is the overall receptive vocabulary size of 4th Spanish Primary school students who are learners of English as a foreign language, as measured by the 1,000 word test and the 2,000 frequency band of the VLT?
b. Will there be significant differences between the vocabulary sizes of male and female students?c.Will there be a significant correlation between students' scores on both the 1,000 word test and the 2,000 frequency band from the VLT and a cloze test?

SUBJECTS
The subjects under study are 270 4th year primary school pupils who are learners of English as a foreign language in four primary schools in La Rioja, Spain.The average age of the students is 10.3 years and the sex distribution is 118 females and 152 males.The sample is homogeneous concerning students' mother tongue, social backgrounds, and type of instruction.All students have Spanish as their mother tongue and attend schools located in middle class areas in a medium size city.According to the language policy of the national and regional governments, the four schools share the same educational goals, similar English language teaching methodology, and an equal number of instruction hours devoted to English.The aim is the achievement of communicative competence by placing emphasis on oral skills with a gradual and integrative introduction of reading and writing skills from the first through to the sixth year (end of primary education).At the time of data collection, the subjects have been taught English for 3 school years in periods of 3 to 4 hours per week for a total of 419 hours.

DATA COLLECTION
Three tests were used in this study: (a) a 1,000 receptive word test (Nation 1993), (b) the 2,000 word frequency band from the receptive version of the VLT (Schmitt, Schmitt and Clapham 2001, version 2), and (c) a sub-test (in cloze format) of a language level test.The selection of the tests was conditioned by the age and the language level of the informants: all three tests have been proved to be within the grasp of young learners such as those found in primary and early secondary education.
According to Schmitt "sampling from the most frequent 1,000 and 2,000 levels is often sufficient, especially for beginners" (Schmitt 2000: 23).For Nation (1993), the most frequent 1,000 words are essential for English language learners because of their high coverage of informal conversations and their presence in English language readers.Likewise, the 1,000 word level test and the 2,000 frequency band of the VLT are drawn from the same frequency lists, which have been used and continue to be used as a basis for many series of graded readers for beginners all over the world.
The cloze test used is a subtest from a standardized language level test for young learners (Corporate Author Cambridge ESOL 2004).This test has been used with young learners of English all over the world and validated as an instrument for discriminating language level.Research has provided evidence of the relationship between the cloze test and vocabulary size test (Jochems and Montens 1988;Fan 2000) as well as the cloze test and language proficiency (Hanania and Shikhani 1986;Lapkin and Swain 1977;Jochems and Montens 1988).Except in Lapkin and Swain, where English and French cloze tests were used to measure children's language proficiency in a bilingual program, most correlational studies have been conducted with adult learners.By using a cloze test, in the present study we aim to ascertain whether this relationship is also observed in primary school learners' scores.The data obtained could serve for a better understanding of the link between vocabulary size and language level and of the relationship between the VLT and the cloze test.

PROCEDURES
The 1,000 word test, the 2,000 frequency band from the VLT, and the cloze test were given to students during class time, within a period of two weeks.They were given 15 minutes to complete each task.At the beginning of each task, clear instructions were given both orally and in written form in the students' mother tongue so as to ensure that they understood what they were being asked to do.To avoid 50% chance of guessing correctly, students were told that wrong answers would be penalized.

RESULTS
Table 2 shows the means and standard deviations for the 1,000 word test and the 2,000 frequency band of the VLT.As can be seen, the mean score for the former is 16.76, whereas for the latter it is 5.33.

DISCUSSION
In research question one, we asked what was the English receptive vocabulary size of 4th Spanish Primary school students as measured by the 1,000 word test and the 2,000 frequency band of the VLT.The results indicate that it falls within the 1,000 frequency level.However, this does not mean that students master this level since scores reveal that half of the students recognize less than two-thirds of the words from this level.Regarding the 2,000 frequency band, the results indicate that few words within this band are known by students.Their profile of receptive vocabulary clearly falls about half way to the mastery of the 1,000 most frequent words (about 737 words). 3In other words, students know 559 words from the 1,000 word test and 178 words from the 2,000 frequency-band of the Vocabulary Level Test.In our view, this finding represents a reasonably good vocabulary size if we consider that the sample of students investigated are still in the fourth year of primary education and that they have been taught English as a curricular subject for three school years and for a total of 419 hours of instruction.Furthermore, this finding is rather positive if we compare it to the results reported in vocabulary size research for secondary school and university students.
A sharp decrease rather than a gradual one is observed from the 1,000 word test to the 2,000 frequency band.On the one hand, this finding shows that the 1,000 and 2,000 word frequency bands are useful in discriminating the students' vocabulary levels.On the other hand, it confirms that 4th primary students' receptive vocabulary profile falls within the 1,000 word band.
As to our second research question, our data demonstrate the existence of a positive correlation between the 1,000 and 2,000 word frequency bands and the cloze test, suggesting a relationship between vocabulary knowledge and L2 proficiency.In this sense, our results confirm the studies carried out by Jochems and Montens (1988), Fan (2000), Hanania and Shikhani (1986), and Lapkin and Swain (1977).As was mentioned earlier, these studies focus on university students in a context of English as a medium of instruction.Our study extends existing research by providing data on the positive correlation of the cloze test and the 1,000 and 2,000 frequency bands of the VLT in the context of English as a foreign language in Spanish primary education.
Finally, in research question three we set out to ascertain whether there are differences in the receptive vocabulary sizes of male and female students.The results show very small although non-significant differences between the two groups.Our results on this point are in line with those found in gender and language education both in L1 and L2 where patterns of difference emerge, but depending on the language aspect analysed, females may outperform males or vice versa.
The results obtained reveal a profile of the receptive vocabulary knowledge of a large and homogeneous sample of English learners in a foreign language context, which we believe will be useful for teachers and researchers.For the former, the pedagogic implications derived from the findings of this study are clear.First, they provide a picture of (the competence) where students of 4th year of primary education are concerning English receptive vocabulary knowledge: the top scores obtained by half of the students in the 1,000 frequency band are 16 to 20 points.This means that in the 1,000 level alone there are almost 500 words that need to be taught.Knowledge of the number of words known in this band by primary school learners has important implications for language education.As Nation (1993: 3)

remarks:
The first 1,000 words of English are essential for all learners who wish to use the language.It is thus very important that teachers know what vocabulary knowledge their learners have and are aware of how they can systematically help them to increase this knowledge.If learners do not know all of the first 1,000 words of English it is well worth ensuring that they have the opportunity to learn those that they do not know.
In the same vein, knowledge of the number of words known and unknown by primary students is relevant for teachers as it allows them to adopt informed decisions on the number of words to be introduced in the lesson as well as the strategies to adopt in the teaching of vocabulary (Read 2000).
Second, since we provide an estimate of how many words 10 year old students in the 4th year of primary education know the results will be useful for comparing learners' estimates of vocabulary size of similar ages and educational level, not only in other parts of Spain but also in other European countries where English is taught in primary schools.Cameron (2003) points out the importance for secondary teachers of receiving information about the young learners who come to them from primary education.This knowledge is essential for constructing the basis of good learning practices.In the same vein, we believe that teachers in the 5th and 6th of primary education need information on their students' receptive vocabulary knowledge in the previous years of language education.As Cameron (2003: 107) remarks: "Taking TEYL seriously involves multiple strands of work, much of which is only just beginning, carrying out new research, learning from programmes across a range of contexts and situations, and understanding more about the nature of child foreign language learning".Third, although differences on vocabulary knowledge in male and females are non-significant they point to the need for awareness of this issue.It may be the case that greater differences are found in older students.In a large-scale study with Spanish secondary students, Jiménez (1992) found significant differences between male and female students.Girls wrote a greater number of words (types and tokens) than boys in English composition tasks.
For researchers, the findings are useful because of the information they reveal with respect to the vocabulary knowledge of a group of learners of which there is very little systematic investigation.The focus of our study on 4th primary school students opens a window on the understanding of receptive vocabulary knowledge of 10 year-olds, an age (according to Piaget) previous to the development of formal thinking.We believe that the data will be particularly useful for comparing vocabulary gains in students' subsequent school years, as well as for studying vocabulary development through time.
This study has focused on the investigation of the receptive vocabulary size of a sample of learners of English as a foreign language at a grade of primary education in a specific region in a concrete specific country of EUROPE.However, although illuminating, the data obtained do not allow us to generalize the results to all 4th Spanish primary students.In spite of the fact that the same official curriculum and similar methodology is followed all over the country, there are bilingual communities in which either Basque, Catalan or Galician are acquired as major or minor languages in addition to Spanish, and where English is taught as a third language.Even in the case of other Spanish monolingual communities, there are differences concerning regional peculiarities and social contexts that may yield different results.Further studies on estimates of students' vocabulary size are needed not only in Spanish bilingual and monolingual communities but also in other countries where English is taught in primary and secondary education in a foreign language context situation.The data obtained would build up a grounded baseline for comparison as well as for unifying vocabulary criteria in the different educational stages all over Europe.
Likewise, care should be taken in the interpretation of the results concerning pupils' receptive vocabulary size.First, the nature of the test used does not allow us to claim more knowledge than word recognition.Although Meara is aware of the limitations of the VLT, he reckons that it is "the nearest thing we have to a standard test in vocabulary" (Meara 1996: 36), Beglar and Hunt observe that this test estimates the learners' basic knowledge of common word meanings (Beglar and Hunt 1999: 132).It does not go further to estimate depth of word knowledge.Second, basic receptive vocabulary knowledge does not guarantee being capable of recalling the words needed or using the words in real communication.More studies are necessary to investigate the validity of the results obtained by means of other measures of the children's English vocabulary knowledge.Likewise there is the need of studying the relationship between receptive and productive vocabulary knowledge as well as the degree of knowledge of the different words contained in the 1,000 word test as well as in the different bands of the VLT.
Finally, although we believe that a profile of the receptive vocabulary of students at a given stage is useful for teachers and researchers, we also believe that it is necessary to carry out longitudinal studies with the same group of learners in order to investigate receptive vocabulary development throughout the different stages of primary and secondary education.Takala (1984) observed that a larger proportion of vocabulary is known at lower stages than at upper stages of education.That is to say, young learners learn more foreign words than older learners.Our guess is that primary students may be more motivated in the learning of the foreign language than secondary students and that motivation may be positively related to vocabulary size, but this is only a hypothesis that requires further investigation.In order to support or disconfirm our hypothesis, and to study the incremental nature of vocabulary acquisition, we plan in the short run to study the development of receptive and productive vocabulary size in 5th, 6th of primary education, and in the long run, it is our intention to investigate the same students in the 1st, 2nd, and 3rd year of secondary education.

Table 1 .
Word source and test format employed in receptive vocabulary studies.