Vol. 8. No. 3 A-1 December 2004

Return to Table of Contents Return to Main Page

Anthony Seow
<tcaseownie.edu.sg>
National Institute of Education
Nanyang Technological University
Singapore

Grace Tay
National Institute of Education
Nanyang Technological University
Singapore

Abstract

This pronoun study examines the effect of two classroom learning environments on the acquisition of English personal and possessive pronouns by Primary Two students in Singapore on the premises that:

1. Students from the formal learning environment will perform better than those from the informal learning environment in the shorter term;
2. Students from the informal learning environment will outperform those from the formal learning environment in the longer run; and
3. Students from both learning environments will find it easier using personal rather than possessive pronouns.

The pre- and post-tests in the study reveal that while both groups of students found it easier to use personal pronouns than possessive pronouns, certain learning outcomes also emerged to suggest that the two groups of learners demonstrated different learning behaviors in pronoun use not only in terms of differentiation in pronoun type but also in pronoun function.

Introduction [1]

This study investigated the instructional effect of formal (i.e., focused on rules and drills) and informal (i.e., communicative) classroom learning environments on Primary Two students' understanding and use of personal and possessive pronouns, namely me, my, mine, you, your, yours, we, our, and ours, on the initial premises that:

1. Students from the formal classroom environment will perform better than those from the informal classroom environment in the shorter term;
2. Students from the informal learning environment will outperform their peers in the formal environment in the longer run; and
3. Students from both learning environments will experience greater ease in using personal pronouns (me, you, we) than possessive pronouns (my, mine, your, yours, our, ours).

Most of the available L2 acquisition studies tend to compare data derived from formal, classroom learning with those from informal, naturalistic (non-classroom) acquisition (see e.g. Ellis, 1991; Krashen, 1976; Larsen-Freeman & Long, 1994; Phillipson, Kellerman, Selinker, Sharwod Smith, & Swain, 1991). Although this pronoun study also examined both formal and informal learning, it was different from most other research studies in that the two types of learning in the pronoun study were observed entirely within the confines of the classroom [2]. In a sense, the informal classroom learning situation simulated a naturalistic, albeit somewhat input-controlled, environment. With the communicative language teaching approach (see Brumfit, 1984; Fotos & Ellis, 1991; Swain, 1985; Widdowson, 1978) replacing or existing alongside the more traditional, formal rule-learning methodology in many language classrooms today, it would be interesting to note the effect of these two competing classroom environments on learning outcomes in young learners.

Supporters of the communicative language teaching approach would argue that comprehensible input even in the classroom (Krashen, 1987) and freely structured opportunities for language use may be sufficient for language acquisition to take place (Tschirner, 1992). What this means is that comprehensible output needs to complement comprehensible input to make acquisition more complete. Fotos (2001, p. 273) observes that "giving learners the opportunity for [comprehensible] output is just as important as giving them input." Put simply, this means that ESL/EFL learners necessarily have to reinforce their knowledge of target-language structures by being sufficiently given meaning-focused communicative activities to give expression to those learned structures or, better still, being actively involved in negotiating meaning when there is real communication breakdown. In the latter instance, there is what Swain (1985, p. 249), in her comprehensible output hypothesis, calls "pushed language use," where the learners are compelled in difficult communicative situations to search for alternative ways of getting their message understood and, in doing so, move from semantic processing during comprehension to syntactic processing during production. [-1-]

In many respects, informal classroom learning, as in the present study, can be said to embrace the communicative methodology (Tayao, 1984) if there is plenty of language production, but is it really more successful than formal learning that advocates a rule-learning orientation (see e.g. Ellis, 1990; Swain, 1985)? Does formal rule-learning, especially for young learners, have a delayed effect? Is there a difficulty order for the learning of English pronouns? These are among the key issues that surfaced in this pronoun study relative to the three premises cited above. The findings could serve as a useful, first tool for prognosticating optimal classroom learning conditions for young learners here in Singapore and elsewhere where similar ESL/EFL classroom learning climate prevails.

Pronoun Research Methodology

Pronoun acquisition research has generated significant interest over the years (Chiat, 1981; Charney, 1980; Felix & Hahn, 1985), although much less of it has been conducted in the area of the personal-possessive pronoun contrast. This is especially true in the local research scene. This study involved a Primary Two class of 32 students from Bedok View Primary School who were given explicit grammar instruction in a formal classroom learning environment, and another Primary Two class of 37 students from New Town Primary School who were exposed to an informal classroom learning environment. These two classes were matched for language aptitude, racial mix, and gender. The students were all seven years of age. Both the classes were equally conversant with the two learning environments but, for the purpose of the pronoun study, followed a specially tailored course of pedagogical treatment of rule-learning and drilling for the formal classroom, and communicative activities among the students for the informal classroom.

The two participating teachers in the two classroom learning environments were given clear, written teaching guidelines to follow. Classroom activities and instructional procedures were also clarified for both the teachers to safeguard against any crossover in methodology, and lesson plans were specially prepared for the two teachers to teach the English pronouns strictly in conjunction with a common learning Unit in the students' Primary Two course book. The types of personal and possessive pronouns covered in the study, and which also appeared in the Primary Two coursebook, are shown in Table 1:

Table 1: Personal and Possessive Pronouns Used in the Study

The students from the formal learning environment were exposed to conscious learning of rules governing pronoun usage through explicit pronoun instruction. The pronouns were explained to the students and students were allowed to practice them in isolation. Repetitive practice, drills, and memorization of pronoun rules were the mainstay of language activities in the formal classroom. Thus, the formal learning environment provided little, if any, opportunity for exploration of authentic contexts for language experience and meaning-driven communication.

In contrast, the students from the informal classroom were involved in show-and-tell sessions and group activities aimed at encouraging the communicative use of the language to facilitate pronoun acquisition. There was no explicit explanation of pronoun rules and the formal aspects of the language. Rather, the students in the informal classroom were guided to understand and use the language globally and intuitively. Before the lesson in each of the two learning environments, the students were given a pre-test to gauge their entry level of pronoun competence. A post-test (Post-test 1) was administered immediately after the lesson to the students in each learning environment to assess the impact of instruction. A month later, a second post-test (Post-test 2) was administered to the same students to establish whether any performance gain identified by the first post-test was sustained. Between the two post-tests, there was no further pronoun instruction. The pre-test and the two post-tests were parallel tests (see Appendix A for a sample test). Each of the pronouns was tested five times in various linguistic contexts in each test. There were forty-five questions on pronoun usage and twenty "dummy" questions, all of which were randomly ordered in each test that covered integrative items in a cloze section as well as discrete-point items in two other sections. Figure 1 shows when the pre- and post- tests were administered:

Figure 1: The Pre-test and Post-tests in both the Formal and Informal Learning Environments

Data Analysis and Results

A frequency count was first made of student performance in each of the three tests--i.e. pre-test, and post-tests 1 and 2--and the results are shown below in Tables 2-4 for the formal learning environment, and in Tables 5-7 for the informal learning environment:

Table 2: Pre-test and Frequency Count in Formal  Learning Environment

Table 3: Post-test 1 and Frequency Count in Formal  Learning Environment

Table 4: Post-test 2 and Frequency Count in Formal  Learning Environment

Table 5: Pre-test and Frequency Count in Informal  Learning Environment

Table 6: Post-test 1 and Frequency Count in Informal  Learning Environment

Table 7: Post-test 2 and Frequency Count in Informal  Learning Environment

It can be seen from Tables 2-4, that the majority of the students from the formal learning environment--that is, 21 out of 32, or slightly over 65% of the students--scored 37 points or above (out of a maximum of 45 points) in each of the three tests. In fact, they did particularly well in Post-test 2, where 21 students managed with 43 points or better. The first hints are that the students in the formal learning environment appeared to take favorably to tests of linguistic forms. This is crucial finding in terms of the input-intake-output debate (see e.g. Schmidt, 1990), which will be discussed further in the next section.

Next, taking the best 24 of the 37 students in the informal learning environment (again, about 65% of the students in this group) for comparison with the earlier group, Tables 5-7 above show that their scores stayed in the 31-45 point region, with slightly better performance demonstrated in Post-test 2 (just like the formal group). However, looking at the overall scores for all of the students in the informal learning environment, it is observed that they were more widely spread out than those from the formal learning environment. This statistic could be understood more clearly in Tables 8 and 9, where the respective means and standard deviations for both groups of students are compared:

Table 8: Means and Standard Deviations in the Formal Learning Environment

Table 9: Means and Standard Deviations in the Informal Learning Environment

Standard deviations for the informal group were consistently larger than those in the formal group in the three tests, indicating that the former group of students was, at least, a more homogeneous group than the latter. Although the two groups came from similar school backgrounds--that is, both schools were neighborhood schools located in government housing localities--the mean test scores of the two groups tended to suggest that the students from the formal learning environment were more competent in the use of English pronouns than those from the informal learning environment, at a particular point in time.

Besides means and standard deviations, the pronoun scores for individual students in each of the two post-tests were also computed in percentage terms for indications of improvement or regression in global performance. For example, students who obtained a higher or a lower score in either Post-test 1 or Post-test 2, when viewed against an earlier test, would be cited accordingly for having shown improvement or regression in that test. As a class, the students' performance, profiling improvement or regression in the post-tests (i.e. by separately totaling the number of students who showed improvement or became worse off in a post-test), is depicted in Tables 10 and 11 below:

Table 10: Rates of Improvement in Test Performance

Table 11: Rates of Regression in Test Performance

From Table 10, we see that the informal class at 54.0% performed marginally better than the formal peers at 46.9%, as reflected in the results of Post-test 1, a test that was taken immediately after pronoun instruction. The data appear to suggest that, contrary to expectations stated in Premise 1 (see the Introduction), the students in the informal learning environment reaped slightly better results than those in the formal environment, at least in the shorter term. Both classes were tested again in Post-test 2 one month later, and test figures show that although the students in both classroom learning environments demonstrated vast improvements in the longer run, the students in the formal classroom learning environment stood to gain much more than those in the informal classroom environment from pronoun instruction in the longer run, as evidenced from the Post-test 2 results. Somewhat counter-intuitively, this finding invalidates Premise 2. [-6-]

In considering Table 11 above, the students in the informal classroom environment were observed to have experienced a regression rate of 27.0% in Post-test 1 (vis-ö-vis the Pre-test). This is lower than that experienced by their formal peers with a regression rate of 40.6% in the same test [3]. An interesting feature of the regression rates seen in Table 11 is that the informal and formal classes showed rapidly decreasing regression rates in the subsequent two post-tests. The decrease was even more remarkable in the case of the formal class, with the students performing to an astounding 0% regression rate in Post-test 2 (vis-ö-vis Post-test 1). All in all, the percentage of students in both classroom learning environments who improved (Table 2) were significantly higher than those who regressed (Table 11). This suggests that the two groups of students generally benefited from classroom lessons in their respective learning environments but at different success rate, with the students in the formal classroom performing better than those in the informal classroom in the longer run.

In the analysis of pronoun sub-types and use, those personal and possessive pronouns that had more than two errors (out of a possible of 5 for each of the pronouns being tested) were considered as "not effectively learned." These pronouns, analyzed thus, were ranked in percentage terms according to the number of students who committed more than 2 errors on each of the pronouns. On this basis, a difficulty order of pronouns acquisition was subsequently derived. Figures 2 and 3 below delineate the three most difficult pronouns to learn, rank-ordered according to the error scores in the Pre-test, Post-test 1 and Post-test 2. [-7-]

[-8-]

One of the most striking findings culled from Figures 2 and 3 above is that, in both the informal and formal learning environments, of all the personal and possessive pronouns examined in the study, the three most difficult pronouns to acquire in each of the three tests were all possessive pronouns. This finding is congruent with Premise 3 in the study. In the informal classroom (Figure 2), the possessive pronoun ours represented the highest number of students who produced the pronoun error in all the three tests and by occupying the top position in the difficulty order for the Pre-test and Post-test 1--with an unusually high number of students in error commission in the latter test--before dropping to second position in Post-test 2. The possessive pronoun yours was a strong contender, sharing the top position with ours in the Pre-test, but witnessed a steady decline in placing to second and third positions in Post 1 and Post-test 2 respectively. The pronoun our fell from second position in the Pre-test to third place in Post-test 1, only to rise again to secure the top position in Post-test 2. In the Pre-test, your made a brief appearance in second place while my and mine came in third, but all these three pronouns disappeared from the difficulty chart after that. What is also clearly evident from Figure 2 is that possessive pronouns with the nominal function (particularly yours and ours) appeared to be the most problematic for the students, and they persisted even after the classroom lesson was delivered, as evidenced in the data from Post-tests 1 and 2.

In the case of the formal classroom (Figure 3), the possessive pronoun ours also had the highest number of students who committed the pronoun error in the Pre-test and Post-test 1 before settling down to second place in Post-test 2. However, unlike that in the informal environment, the possessive pronoun yours saw a rise in the number of students who produced errors in this pronoun: it rose steadily from third position in the Pre-test to second in Post-test 1 and then to the top spot in Post-test 2. The pronoun mine remained in third position for the two post-tests after dropping from second place in the Pre-test, while our occupied third position for all the three tests.

In each of the two learning environments, certain types of pronouns are obviously more difficult to acquire than others. In averaging the number of errors observed in Post-test 1 and Post-test 2 for all of the pronouns that were produced by the students in both the learning environments consequent upon classroom instruction, some telling statistics emerge, as seen in the following Figures 4 and 5:

[-9-]

It is evident from Figures 4 and 5 that possessive pronouns were more difficult to acquire than personal pronouns for the students in the informal classroom as well as in the formal classroom. This finding confirms the initial hunch that personal pronouns rather than possessive were less likely to create learning difficulties for both groups of students (Premise 3).

Except for my in the formal classroom (see Figure 5 above), the average number of errors in Post-test 1 and 2 for each of the possessive pronouns in the two learning environments was higher than that for personal pronouns. Another noteworthy finding is that possessive pronouns with the nominal function (i.e. mine, yours, ours) were more problematic than possessive pronouns with the determiner function (i.e. my, your, our) for the students in both learning environments. The only exception is mine which had marginally fewer errors than my produced by the students in the informal environment (see Figure 4 above).

In the study, pronoun use was also examined in terms of the actual number of errors produced for each of the pronoun types. In tallying pronoun errors thus, Table 12 below offers another view of the way the students in both learning environments managed pronoun use before (as evidenced in the Pre-test) and after their class lesson (as observed in Post-tests 1 and 2). What is immediately apparent in Table 12 is that the number of errors in the Pre-test and the two post-tests was consistently and substantially lower for each of the pronoun types in the formal classroom environment than that in the informal environment. It suggests that, because of the comparatively fewer errors made, the students in the formal classroom could be said to be generally more skillful at pronoun use than their informal counterparts, after classroom instruction.

While initially Post-test 1 results show that, generally, the informal students tended to have benefited more from classroom instruction in that they made fewer errors across the spectrum of pronouns (i.e., in me, my, mine, your, yours and our) than those formal students who, in the same post-test, managed to gain from classroom instruction on only three pronouns (namely, my, mine and ours). However, in Post-test 2, although both formal and informal students showed improvement in all but one pronoun, the gain, especially in percentage terms (see Table 13), was more remarkable for the formal group than for the informal group of students, indicating that in the longer term, the formal students indeed appeared to benefit more from explicit grammar teaching.

Table 12: Total Number of Pronoun Errors in Pre-test, Post-test 1 and Post-test 2

Table 13: Comparing Pronoun Errors in Pre-test and Post-test 2: Effect of Instruction in the Longer Run

Discussion of Findings and Implications for Teaching

It is clear from the pronoun study that both classroom learning environments are conducive to the learning of pronouns, although in different ways. Despite the fact that the formal class was generally more competent in pronoun use than the informal group, the formal class did not seem to benefit immediately from direct instruction although they managed to outperform the informal class by having fewer pronoun errors in the longer term. One plausible explanation for this is that there could be a delayed effect of instruction (cf. Lightbown, 1987). It seems that especially for very young learners (7-year-old Primary Two students in the formal learning environment, in our case), explicit instruction in grammar could have initially heightened their level of linguistic awareness--not unlike "consciousness-raising," in Sharwood-Smith's (1981) understanding--about the existence of certain grammatical differences in pronoun types and functions and, although "noticing" (Schmidt, 1990) resulting in "deep retention" (Chaudron, 1985) of pronouns was not immediately apparent following explicit instruction, through a longer exposure to the target language the formal learners were able to sort out the rules of pronoun usage for themselves. Here, Swain's (1985) comprehensible output hypothesis, which could be observed to complement Krashen's input hypothesis (1981), also has relevance. What this present pronoun study might suggest is a restatement of the relations of input, intake and output, as exemplified in Figure 6 below:

Figure 6: Input, intake and output

¹Krashen (1981)
²Chaudron (1985)
³Chaudron (1985)
⁴Gass (1988); Schmidt (1990); Sharwood Smith (1986); Swain (1985)

The re-statement of the input-intake-output hypothesis may hint that input is a crucial but not a sufficient condition for the acquisition of grammatical accuracy to occur; a learner would need "the opportunity for meaningful use of her linguistic resources to achieve this" (Ellis, 1991, p. 171). From Figure 6, we may intimate that not all available input is comprehensible to the learner. However, that which is indeed "comprehensible input" (Krashen, 1981) to the learner would constitutes what Chaudron (1985) calls "preliminary intake" which accounts for all the stored data that the learner comprehends in order to process language. Ellis (1991), in echoing Swain (1985), sees that eventually "production, as opposed to comprehension, may help to force the learner to move from semantic to syntactic processing", clarifying that "it is possible to comprehend a message without any syntactic analysis of the input it contains" (Swain, 1985, p. 117). What this means is that "comprehensible output" or, more precisely, "output that extends the linguistic repertoire of the learner as he or she attempts to create precisely and appropriately the meaning desired" (Swain, 1985, p. 252) must have stemmed from "final intake" (Chaudron, 1985), which is itself a part of the larger corpus of comprehensible input responsible for enabling the learner to understand and use stored linguistic data in the first instance. [-11-]

If all that is thus far argued is tenable, then we may posit that classroom instruction in this pronoun study appears to have a different impact upon the formal and informal groups of learners. Both groups gained from input presented differently, with the formal group focused more on rule formation and the informal on interaction. Initially, at least, both groups would have to rely on comprehensible input, however presented, to comprehend and use pronouns, and for this--as Post-test 1 results have shown--the students in the informal learning environment tended to have benefited more from the communicative activities than those in the formal learning environment. However, given time, the formal students outpaced the informal counterparts (as seen in the Post-test 2 results) probably because the former were forced or "pushed" much harder to "notice" and use the linguistic forms formally taught to them. Put differently, while both groups of students would have acquired sufficient final intake in separate ways for differential comprehensible output to occur, the formal groups were observed to outperform the informal group partly because the nature of the objective pronoun tests (i.e., the two post-tests) favored those students in the formal learning environment who were, as it were, predisposed at the outset to focus more on form. [4]

However, this is not to say that we should all abandon the informal classroom learning in favor of formal learning. Formal teaching of grammar in the classroom if done to the exclusion of communicative methodology that is normally seen in most ESL learning environments may not always produce the desired results (see e.g. Ellis, 1991, pp. 134-135). Prior to the present study, both groups of students had been accustomed to the two kinds of learning environments. Owing to the interactive nature of the classroom lesson, the informal students had, in fact, the initial advantage of performing better than the formal students who were still trying to come to grips with the new pronoun rules of usage. The implication of all this for ESL teaching is that it would do well to complement the mainly implicit learning by learners with occasional strong doses of explicit teaching of grammar, since the informal classroom learning environment while highly desirable may not in itself be adequate for full pronoun acquisition to occur. Explicitly taught grammatical knowledge seems to play a part to help learners notice and retain the organized linguistic systems that, in turn, facilitate the acquisition of grammatical accuracy.

A more significant outcome of this pronoun study is that possessive pronouns are consistently more difficult for both groups of students to acquire, and that, within the possessive pronouns, those with the nominal function (e.g. mine, yours, ours) are generally more problematic than those with the determiner function (e.g. my, your, our). This could be attributed to the fact that, in the scheme of English pronouns, there could well be a developmental order for learning personal and possessive pronouns. While this could be substantiated by further research, at least for the moment, we could tentatively proffer the following stages of pronoun learning in the classroom, based on the analysis of student errors in the present study:

Stage 1:
Personal pronouns (e.g. me, we, you)

Stage 2:
Possessive pronouns with determiner function (e.g. my, our, your)

Stage 3:
Possessive pronouns with nominal function (e.g. mine, ours, yours)

Personal pronouns (apart from you) have clearly distinctive surface forms, and thus learners are less likely to confuse form with function or use. In contrast, possessive pronouns in their two functions are likely to be more confusing to the learners especially in terms of their closeness of surface form and/or pronunciation in the my-mine, your-yours and our-ours contrasts. Thus, Stage 3 pronouns could be seen as a natural progression from Stage 2 in the following respect:

Stage 2
This is our dog.

Stage 3
This dog is our dog.
This dog is ours.

In the above linguistic context, semantically, our is equivalent to "our dog." The same deductive procedure could be ascribed to the other two pairs of possessive pronouns.

Methods in language teaching come and go. Richards agreeing with Mackey (1965), believes:

[A]lthough there has been a preference for particular [teaching] methods at different times, methods often continue in some form long after they have fallen out of favour; this observation is still true today, with the grammar translation still alive and well in some parts of the world. (2001, p. 3)

This perhaps explains why the explicit teaching of grammatical rules--as exhorted by the previously popular structural syllabus--and the occasional use of drills are not altogether looked upon with disfavor in the face of the currently more popular communicative methodology in most classrooms today, and willy-nilly this method of instruction for pronoun acquisition or some other is likely to continue for some time yet in the ESL setting.

This pronoun study suggests that, for very young learners, while personal pronouns and possessive pronouns with the determiner function may be taught early in the course, it might be best to defer the teaching of possessive pronouns with the nominal function until a later stage. Additionally, for the young ESL learners, the transfer of explicit pronoun learning for language production may not be immediate, suggesting that formal pronoun rules per se may not mean very much to the learners unless, in negotiating meaning in communicative situations, they notice the very same pronouns used meaningfully over a period of time that is long enough for acquisition to occur. [-13-]

Notes

[2] A distinction has often been made between conscious learning of the rules of a target language in a formal classroom setting, and subconscious acquisition of the target language in an informal, naturalistic environment where the language is merely picked up without any teacher intervention (Krashen, 1976).

[3] In examining Tables 2 and 3, it should be noted that while these were students who demonstrated improvement or regression, there were others who showed no performance change at all. Thus, if 27.0% of the informal class regressed in Post-test 1 since the Pre-test (see [d] in Table 3), and another 54.0% of the same class improved (see [a] in Table 2), it could be inferred that 19.0% of the informal class actually did not improve or regress; that is, they showed no change at all in performance.

[4] It would be interesting to find out whether, with tests involving spontaneous speech or writing in a classroom setting, similar performance behaviors would prevail for these two groups of young learners.

About the Authors

M.H. Grace Tay, was formerly a BA (Honours) trainee-teacher at the National Institute of Education. She now teaches in a primary school in Singapore. [-14-]

References

Brumfit, C. (1984). Communicative methodology in language teaching. Cambridge: Cambridge University Press.

Chiat, S. (1981). Context-specificity and generalization in the acquisition of pronominal distinctions. Journal on Child Language, 31, 75-91.

Charney, R. (1980). Speech roles and the development of personal pronouns. Journal of Child Language, 7, 509-528.

Chaudron, C. (1985). Intake: on models and methods for discovering learners' processing of input. Studies in Second Language Acquisition, 7(1), 1-14.

Ellis, R. (1990). Understanding second language acquisition. (6th ed.). Oxford: Oxford University Press.

Ellis, R. (1991). Instructed second language acquisition. Oxford: Basil Blackwell.

Felix, S., & Hahn, A. (1985). Natural processes in classroom second language learning. Applied Linguistics, 6(3), 223-238.

Fotos, Sandra (2001). Cognitive approaches to grammar instruction. In M. Celce-Murcia (Ed.), Teaching English as a second or foreign language (3rd ed.) (pp. 267-284). Boston, MA: Heinle & Heinle.

Fotos, S., & Ellis, R. (1991). Communicating about grammar: A task-based approach. TESOL Quarterly 25(4), 605-628.

Gass, S. (1988). Integrating research areas: a framework for second language studies. Applied Linguistics, 9, 198-217.

Krashen, S. (1976). Formal and informal linguistic environments in language acquisition and language learning. TESOL Quarterly, 10, 157-168.

Krashen, S. (1981). Second language acquisition and second language learning. Oxford: Pergamon Press.

Krashen, S. (1987). Principles and practice in second language acquisition. Englewood Cliff, N.J.: Prentice Hall International.

Larsen-Freeman, D., & Long, M. (1994). An introduction to second language acquisition research. London: Longman.

Lightbown, P. (1987). Classroom language as input to second language acquisition. In C. Pfaff (Ed.), First and second language processes (pp. 169-187). Rowley, Mass.: Newbury House.

Mackey, W.F. (1965). Language Teaching Analysis. London: Longman.

Ministry of Education (2001). Guide to the English language syllabus 2001: Primary 1-4. Singapore: Curriculum Planning & Development Division.

Phillipson, R.,Kellerman, E.,Selinker, L., Sharwood Smith, M., and Swain, M. (1991). (Eds.), Foreign/second language pedagogy research. Philadelphia: Clevedon.

Richards, J. (2001). Curriculum development in language teaching. New York: Cambridge University Press.

Schmidt, R. (1990). The role of consciousness in second language learning. Applied Linguistics, 11(2), 129-158. [-15-]

Sharwood Smith, M. (1981). Consciousness-raising and the second language learner. Applied Linguistics, 2, 159-169.

Sharwood Smith, M. (1986). Comprehension versus acquisition: two ways of processing input. Applied Linguistics, 7, 239-256.

Swain, M. (1985). Communicative competence: some rules of comprehensible input and comprehensible output in its development. In S. Gass, & C. Madden (Eds.) Input in second language acquisition (pp. 235-256). Rowley, MA: Newbury House.

Tay, Meng Hon Grace (1977). The effect of two classroom learning environments on pronoun acquisition among Primary Two students. Unpublished BA (Hons) Academic Exercise. Singapore: National Institute of Education, Nanyang Technological University.

Tayao, M. (1984). Turning the classroom into an acquisition-rich setting: A review of related research. RELC Journal, 15(1), 97-107.

Tschirner, E. (1992). From input to output: Communication based teaching techniques. Foreign Language Annals, 25, 507-518.

Widdowson, H. (1978). Teaching language as communication. Oxford: Oxford University Press.

To the appendices