Syntactic Proficiency of Kuwaiti EFL Learners : A Study of Tag Questions

This study examines the extent to which 70 Kuwaiti EFL learners are aware of the formation rules of tag questions in English. It also investigates whether the English proficiency level of the participants contributes to their correct answers on the test. For this purpose, the researchers used a multiple-choice test to measure the participants’ awareness of the correct form of tag questions. The results show that Kuwaiti EFL learners are not aware of the syntax of tag questions in English (total mean = 47%). The results also reveal that the English proficiency level of the participants plays an important role in their correct answers on the test; there arestatisticallysignificant differences between the results of the Advanced Learners (ALs) and the Intermediate Learners (ILs). The researchers note that the participants encounter difficulties with certain types of tag questions, especially irregular canonical tag questions. The difficulties are ascribed to many reasons such as overgeneralizationof the formation rule, partial application of the formation rule, lack of subject-agreement, wrong verb form and misconception of the polarity rule. Essentially, the non-existence of tag questions in the participants’ first language i.e. Arabic may be the most prominent reason for the participants’ inability to provide accurate answers on the test. The study concludes with some pedagogical implications that may assist teachers of English as a second/foreign language to highlight the correct use of tag questions in English in the classroom.


Introduction
Tag questions are usuallyused as hedging devices in English; they are used when the speaker is not certain about what he/she has heard, or when he/she wants to verify the truthfulness of a particular statement.As commonly known, tag questions are usually added to statements (i.e., a declarative sentence or an imperative) to give them the structure of a question.The declarative portion is then inverted by adding the tag question.The statements can have a positive or negative polarity.However, this polarity is reversed from the statement to the tag.Examples 1-5 show some instances of such interrogative clauses: 1) She is smart, isn't she?
2) He went to the party last night, didn't he?
3) Answer the phone, will you?4) I am going to prison, aren't I? 5) Nobody wants to go to jail, do they?
The use of these devices is quite pervasive amongst speakers of English, which makes their acquisition a necessity for EFL learners.The mastery of such devices may help EFL learners sound more-native like; thus, enhance their self-confidence, especially when conversing with native speakers of English.Due to their complex syntactic, semantic and prosodic structure, EFL learners, in general, encounter many obstacles in learning English tag questions and using themproperly (Al-Nabtiti, 2012).It has been argued that mastering the use of tag questions in English requires an advanced proficiency level; the competent use of these devices requires considerable communicative skills (Holmes, 1982).This study aims to test the extent to which Kuwaiti EFL learners are aware of tag questions in English.Specifically, it investigates whether the English proficiency level of the participants plays a role in theirawareness of tag questions in English.The ultimate goal is to shed more light on the acquisition of different aspects of the English language by EFL learners and the factors that may contribute to this process.

Tag Questions in English
According to Holmes (1982, p. 41), tag questions can be classified into two types: 1) grammatically complex tag questions or canonical tag questions, an example of such a type is John has lost his wallet, hasn't he? and 2) grammatically simple tag questions or invariant tag questions e.g., Jennifer is silly, right?.To form the former type, the speaker has to pay attention to four principles: 1) the same auxiliary of the statement is used in the tag question; however, if there is no auxiliary in the statement, the auxiliary do is used e.g., she left, didn't she?; 2) the subject of the tag is the same as the one used in the statement e.g., he is going to France, isn't he?; 3) the polarity of the statement is the opposite to that of the tag that is, if the statement is negative, the tag is affirmative and vice versa e.g., my mother is sick, isn't she?, Layla has never lied to you, has she? and 4) the tone falls on the auxiliary or on not if used in the full form, the tone can be either rising or falling (Quirk et al., 1985, p. 810).
In addition to their syntactic complexity, tag questions usually serve an array of social/pragmatic functions.For instance, tag questions can be used to obtain information in the form of a polite request uttered with rising intonation e.g., Jennifer is still in the office, isn't she?This function is referred to as informational tag by Alego (1988Alego ( , 1990)).In addition to informational tags, four other functions can be identified: 1) confirmatory tags, which are usually uttered with falling intonation, are used by the speaker to ask the addressee for support regarding his/her statement e.g., we are going to the hotel tonight, aren't we? 2) Peremptory tags are usually used to end a debate or conversation on a certain topic e.g., we should leave it here, shouldn't we? 3) Punctuational tags request the addressee to pay attention to what the speaker is talking about e.g., I've just mentioned this issue, haven't I? and 4) Aggressive tags, which are usually uttered with a falling intonation, are used to accuse the addressee of being unreasonable or lacking good will e.g., you're stupid, aren't you?(Alego, 1988;1990).
Taking the complex syntactic structure of tag questions into consideration on the one hand, and the social/pragmatic functions performed by them on the other, one may argue that their acquisition by EFL learners isconsidered a stumbling block.The next section reviews some studies conducted on the acquisition of tag questions by children and EFL learners.

Studies on the Acquisition of Tag Questions
The acquisition of tag questions has received a prominence in the research concerned with child language acquisition.For example, Dennis et al. (1982) investigated the acquisition of English tag questions by fifty children whose ages ranged between 6-14 years old.Particularly, they tested whether the children are able to produce tag questions.The results revealed that the children's ability to produce tag questions improved from six to eight years old; however, no such improvement was detected thereafter.The results also showed that the acquisition of the rules required to form tag questions, namely, using the same subject, verb form, polarity as well as inversion varied according to the child's age.Specifically, the polarity rule was mastered by half of the participants whose ages ranged between 10 and 14, whereas the inversion rule was mastered by half of the younger participants.The researchers concluded that the linguistic skills which involve simultaneous manipulation of different syntactic features on the surface level are usually acquired in later stages of language development.
In a similar vein, Weeks (1992) investigated preschool children's ability to repeat positive or negative polarity tag questions.To determine whether the patterns of repetition differ between children of low language skills and those of normal language skills, Weeks (1992) also tested whether the children conformed to the polarity principle when they produce tag questions.The results revealed that the polarity principle was adhered to by children of high language level as opposed to those of low language level.However, both groups violated the polarity principle by producing tag questions whose polarity matched that of the preceding statement.This result may demonstrate that children do not always adhere to the input they receive from their parents or other individuals; they sometimes produce faulty structures that they may have never heard before.
Other researchers examined EFL learners' awareness of tag questions in English.For example, Schaffer (2002) examined Korean EFL learners' ability to respond to tag questions in English.Based on the analysis of their responses, Schaffer (2002) provided insights into the ways by which the culture affects the language in Korea, especially concerning responses to tag questions.In particular, Schaffer (2002) explained that in the Korean culture, it is considered improper to disagree with another.Basically, the polite and honorific nature of Korean affects their responses to tag questions, especially the way they are formed.In addition, the results showed that the erroneous responses produced by Korean speakers of English involved the violation of the negative agreement rule, which requires the use ofno and don'tin the negative response, and yes and do in the affirmative one.Schaffer (2002) explained that negative L1 transfer affects learners in that they will usually depend on invariant tags in the target language due to their similarity to Korean tag questions.L1 negative transfer can also be noted in the inaccurate responses supplied by Korean EFL learners when their response to the tag should be negative.This is because Korean tag questions produce positive responses.Essentially, Schaffer (2002) concluded that L1 negative transfer can be considered the main reason behind the participants' faulty answers.
In another recent study, Al-Nabtiti (2012) investigated the obstacles that Arab ESL learners face when producing English canonical tag questions.The researcher used production tasks to measure the participants' ability to form correct tag questions.The sample of the study included two groups; the first group consisted of thirty-two adult ESL learners, who lived in Canada, whilst the second group comprised ten native speakers of English.The reason for including the latter sample was to compare the performance of the former group with that of the latter on the tests.The results revealed thatESL learners were not able to use the canonical tag questions in an appropriate manner.The reason for this difficulty was not attributed to the syntactic complexity of English tag questions, rather to their pragmatic requirements.Hence, the researcher concluded that ESL curricula need to be amended that is, pragmatic theory and the social context in which the discourse occurs need to be incorporated and explained properly.
By examining the literature review on the acquisition of tag questions by EFL learners in English, one may note that the acquisition of tag questions by second/foreign language learners has not received much attention in general.The focus has been directed to child language acquisition.Therefore, it can be proposed that more studies need to be conducted to tackle this issue.Hence, this study seeks answers to the following questions: 1) To what extent are Kuwaiti EFL learners aware of the correct use of canonical tag questions in English?
2) To what extent does the English proficiency level of Kuwaiti EFL learners influence their acquisition of English tag questions?
3) What are the main sources of difficulty (if any) encountered by the participants on the test?

Research Hypotheses
The researchers formalised the three following hypotheses and set out to test their validity: 1) Kuwaiti EFL learners are not aware of the correct use of canonical tag questions in English.
2) There are statistical significant differences between the results of ALs and ILs on the test.
3) Some types of tag questions are more problematic to the participant than others.
The next section provides an overview of the methodology.

Sample
The sample included seventy students at the Public Authority of Applied Education and Training (PAAET) and College of Basic Education (CBE).Their mean age was 22 years old, and they were all native speakers of Kuwaiti Spoken Arabic (KSA).In order to ensure the validity of the results, the participants were selected based on simple random sampling out of approximately 8000 students.As the participants' English proficiency level was an independent variable, the participants were selected based on their scores on the English Placement Test (EPT).Particularly, the participants who scored 50-69 on the EPT were considered intermediate, whereas those who scored 70 -85 were considered advanced.The seventy participants were divided into two groups: 1) thirty five Intermediate Learners (ILs); and 2) thirty five Advanced Learners (ALs).The researchers opted for intermediate and advanced learners based on their belief that participants with low English proficiency level may not pass a test on structures that are syntactically complex (Holmes, 1982).As a result, participants with low English proficiency level were excluded from this study.It is worth noting that the participants took part in this study voluntarilyso that, only those who agreed to be involved were included.

Instruments
The multiple-choice test is one of the most well-known and widely used assessment tools for measuring the participants' comprehension of a particular issue (Nicol, 2007, p. 54).Thus, the researchersselected the multiple-choice test based on their belief that it is the appropriate method to measurethe participants' ability to select the correct tag question in English based on the preceding statement.The test was designed based on the model originally proposed by Charteris-Black (2002, p. 119), it included 20items (see Appendix A).The choices on the test were designed as follows: one correct answer (the correct tag), one wrong answer (reversed polarity of the correct tag), one distracter (wrong form of the tag), and one I don't know option.The last option was used to reduce the probability of selecting the wrong answer, which in turn will ensure the validity of the test.

Statistical Analysis
In order to provide accurate results, the researchers used the Statistical Package for Social Sciences (SPSS).In particular, the statistical analysis included the calculations of means and standard deviations of the answers of the two groups who participated in the study.To test whether there are statistical significant differences between the results of the two groups, the researchers used a t-test, which compares two means and determines whether the differences between them are statistically significant.

Results and Discussion
As mentioned previously, this study aims to test whether the English proficiency level of seventy Kuwaiti EFL learners plays a role in their acquisition and use of tag questions in English.Additionally, it aims to investigate the main sources of difficulty and account for them.The results showed that the total mean of correct answers for both groups was 47%, which means that Kuwaiti EFL learners are not aware of the syntax of tag questions in English.Put differently, the first hypothesis was confirmed.Table 1 shows the results of the t-test.  1 clearly shows that the English proficiency level of the participants plays an important role in their results on the test.Specifically, there is a statistical significant difference between the results of ALs and ILs;the statistical significance (0.023) is lower than (0.05).This means that the second hypothesis was confirmed.Table 2 shows the results of ALs and ILs on some items in which ALs scored higher than their ILs counterparts.Total mean % 88% 49% 69% Table 2 demonstrates that ALs provided a higher percentage of correct answers (total mean = 88%) than their ILs counterparts (total mean = 49%) on regular canonical tag questions.Possibly, such a type of tag questions required less effort than the other types discussed in this section.Specifically, the participants had to match the subject and the auxiliary of the tag with those of the preceding statement and reverse the polarity.This competence was much more evident in ALs answers as opposed to those of ILs.Therefore, the difference between the answers of both groups was statistically significant.Despite their good performance on regular canonical tag questions, ALs overall mean shows that they still faced substantial difficulties with other types of tag questions.Table 3 shows the total means of the participants' answers on the test.An examination of table 3shows that the Advanced Learners (ALs) outperformed their Intermediate Learners (ILs) counterparts on the test in general.In particular, the total mean of correct answers of ALs (62.4%) is higher than that of ILs (31.4%).This percentage shows that ALs is more aware of the correct syntactic form of tag questions in English.However, the total mean of both groups ALs and ILs (47%) demonstrates that the participants are not generally aware of the syntax of tag questions in English.
Through careful study of the data, the researchers noted that the participants in both groups encountered difficulties in answering certain types of tag question.For instance, the participants found the tag questions whose verb form is different from the one in the preceding statement particularly difficult.Table 4 shows the participants' results on such items.A careful study of table 4 shows that both groups found this type of tag questions quite difficult (total mean of both groups = 35%).As mentioned previously, canonical tag questions are quite complicated, especially from a syntactic viewpoint.The participants needed to take into account the gender and number of the subject of the preceding statement on the one hand, and the verb tense, form, and polarity on the other.In other words, the verb used in the tag should agree with that found in the preceding statement.Taking these factors into consideration, the irregular type of tag questions described in Table 4 may have presented a bigger challenge to the participants of both groups.Several participants provided the following answers on the test: 1) *She really needs new pants, needn't she?
2) *Jennifer has a mole on her left cheek, hasn't she?
3) *We had fun last night at the party, hadn't we?
4) * I'm funny,*am not I? (Note 1) One may argue that these erroneous responses can be attributed to overgeneralising the formation rule of tag questions.This occurs when the participants wrongly apply the formation rulewhich requires the auxiliary in the tag to be identical to that of the preceding statement.As a result, the participants choose the wrong form since the verbs found in the preceding statements in the above examples are main verbs, not auxiliaries.Thus, they providefaulty answers.
Other instances taken from the participants' answers showed that many ILs had a misconception about the formation rule of tag questions in English.In particular, they possibly thought that canonical tag questions in English are always negative, regardless of the polarity of the preceding statement.As a result, they supplied inaccurate answers on the test.For instance, 34% of ILs provided the following answer: *Mary didn't go to school yesterday, didn't she?
Through their examination of the participants' results, the researchers noticed that the participants found the statements which contain expressions such as never, nobody and nothing challenging.Table 5 shows such tags which were extracted from the test (see Appendix A).A look at Table 5 shows that this type of tag questions elicited a big number of wrong answers of the test (total mean of answers = 31%).The main source of difficulty stems from the presence of never, nobody and nothing in the preceding statement.Possibly, the participants were not able to work out the correct form of the tag since the aforementioned words denote a negative sense, without the presence of an explicit negative form not. Therefore, the participants may have not been able to reverse the polarity of the preceding statement; thus, they provided inaccurate answers on the test.For instance, several participants provided the following answer: 5) *John never eats nuts, doesn't he?
This inability to provide the correct tag could be ascribed to partial application of the formation rule.With regard to tag questions that included nobody and nothing in the preceding statement, the challenge was twofold.Firstly, the participants had to work out the correct polarity of the tag since nobody and nothing denote a negative sense, the tag has to be affirmative.Secondly, they had to pay attention to subject-agreement between the preceding statement and the tag.Thus, the participants had to exert much effort to arrive at the correct form of the tag.
Furthermore, the data analysis showed that the participants faced considerabledifficulties with imperative statements, and those that were subjectless.Table 6 shows the participants' answers on such statements.A careful look at table 6demonstratesthat the participants from both groups encountered immense obstacles with imperative statements and subjectless ones (total mean of correct answers = 25%).This type of tag questions was particularly difficult to the participants for two reasons.Firstly, the participants had to exert much effort to find the correct form of verb that should be used in the tag, given the fact that the preceding statement does not provide them with any clues about the correct form of the verb.Secondly, they needed to discover the subject that should be used, since the preceding statement offers no hints about the subject required in the tag.
Essentially, the researchers may argue that the main source of difficulty in forming tag questions in English stems from the non-existence of tag questions in Arabic.Therefore, it has been noticed that EFL learners usually avoid using canonical tag questions, and prefer to use invariant tag questions instead due to the syntactic complexity of the former.The participants' inability to provide the correct form of verb and to reverse the polarity of the preceding statement may be considered a direct result of the non-existence of such a structure in Arabic.

Conclusion and Recommendations
This study examined the extent to which seventy Kuwaiti EFL learners are aware of canonical tag questions in English.Specifically, it investigated whether the English proficiency level of the participants contributes to theircorrect answers on the test.It also attempted to account for the main sources of difficulty which the participants encounteredin comprehending tag questions.The results revealed that Kuwaiti EFL learners are not aware of the syntax of tag questions in English (total mean = 47%), in other words, the first hypothesis was confirmed.They also showed that the participants' English proficiency level influenced the participants' correct answers on the test.ALs (total mean of correct answers = 62.4%) outperformed their ILs (total mean of correct answers = 31.4%)counterparts on the test.This means that the second hypothesis was confirmed.The results also showed that the participants encountered difficulties with certain types of tag questions, mainly, irregular tag questions whose verb form is different from that of the preceding statement, tags that contained never, nobody and nothing in the preceding statement and tags preceded by imperative/subjectless statements.The main sources of difficulties were attributed to: 1) overgeneralization of the formation rule of tag questions; 2) partial application of the formation rule; 3) lack of subject-agreement between the preceding statement and the tag; 4) wrong verb form; 5) misconception of the polarity rule; and 6) lack of awareness of tag formation rules in imperative/subjectless statements.The non-existence of tag questions in the participants' first language i.e., Arabic may be regarded as the most obvious reason for the participants' inability to provide accurate answers on the test.In sum, the results showed that the third hypothesis was also confirmed.
Taking the results into consideration, teachers of English as a second/foreign language need to pay attention to tag questions and their correct formation rules in the classroom.The teacher can attempt to expose the students to everyday English language in order to highlight the contexts in which tag questions are used, and the social/pragmatic functions for which tag questions are produced.In addition, students need to be provided with opportunities to use tag questions orally during class.This can enhance their ability to produce tag questions, which in turn improve their English communication skills.Also, teachers need to provide feedback to the students on their use of tag question in oral exercises during class.This feedback can promote students' performance and help them remember the correct form of tag questions.The results elicited from this studymayhelp inform teachers of English as a second/foreign language on new methods that can be followed during class to teach tag questions, such as engaging the students in role-play exercises that encourage them to use tag questions in English.The researchers recommend that more studies need to be conducted on the awareness of EFL learners of different aspects of the English language, and the methods by which their performance can be enhanced.

Appendix A
The test Level: _____________ Q1: choose the best answer to complete the following sentences.
has a mole on her left cheek, ___________?a) doesn't she b) hasn't she c) does she d) I don't know 6) Bring me a glass of water, __________?a) should you?b) will you c) shouldn't you d) I don't know 7) Layla was crying yesterday, __________? in China can make noodles, ___________?a) can they b) do they c) can't they d) I don't know 19) Nobody knows about his affair, _________?a) does she b) do they c) don't they d) I don't know

Table 1 .
Results of t-test of differences between (ALs) and (ILs)

Table 3 .
Total mean of the correct answers of (ALs) and (ILs) on the test

Table 4 .
Results of (ALs) and (ILs) answers on tags whose verb form is different from that of the preceding statement

Table 5 .
Results of (ALs) and (ILs) answers on tags that contained never, nobody and nothing in the preceding statement