Articles
What Type of Knowledge does the TOEIC Bridge Test Measure?
2021 Volume 5 Pages 1-16
Details
2021 Volume 5 Pages 1-16
This study aims to examine whether the TOEIC Bridge Test measures the declarative and/or procedural knowledge of L2 learners. Elder and Ellis (2009) investigated the relationships between two standardized proficiency tests (TOEFL and IELTS) and the implicit and explicit knowledge of L2 learners. Their results suggest that different L2 proficiency tests may encourage the use of different types of knowledge. However, no previous research has investigated the relationship between the TOEIC Bridge test and declarative/procedural knowledge, which this study undertakes. The results indicate that although the TOEIC Bridge test assesses both declarative and procedural knowledge, the relationship between procedural knowledge and the TOEIC Bridge test was relatively stronger than that between declarative knowledge and the TOEIC Bridge test. In addition, the results suggest that the reading section of the test might be more related to declarative knowledge than the listening section. Thus, these results support Elder and Ellis’s argument that different standardized L2 proficiency tests may measure different types of grammatical knowledge.
In English as a foreign language (EFL) and English as a second language (ESL) contexts, standardized English proficiency tests have been used to measure second language (L2) learners’ proficiency. The scores obtained from such tests are used for multiple purposes such as grouping students based on their scores and planning lessons that consider L2 proficiency level. L2 learners may sometimes need to achieve a high score in the tests to enter a university, receive credit from a university, or even get the job that they want. In addition, if they study the L2 diligently and earn a higher score each time they are tested, they will gain confidence in their L2, which could motivate them to continue learning. However, what type of knowledge do the proficiency tests actually measure? Elder and Ellis (2009) investigated the relationships between the L2 proficiency scores obtained from standardized tests, such as the Test of English as a Foreign Language (TOEFL) and the International English Language Testing System (IELTS), and students’ implicit (procedural) and explicit (declarative) knowledge. However, the relationships between the Test of English for International Communication (TOEIC) Bridge and declarative/procedural knowledge have not been examined. The TOEIC Bridge test is used in some high schools and universities (The Institute of International Business Communication, 2019). Therefore, clarifying these relationships is of importance not only for researchers but also for second language teachers and learners. For instance, L2 teachers might be able to infer the extent to which their teaching can affect their students’ declarative/procedural knowledge based on the scores.
1.1 Declarative and Procedural KnowledgeDeKeyser (2015) stated that L2 knowledge develops from declarative knowledge through procedural knowledge to automatized procedural knowledge. Declarative knowledge is “knowledge of facts (semantic memory) and events (episodic memory); usually consciously accessible and often verbalizable, but not necessarily” (DeKeyser, 2017, p. 17). In contrast, procedural knowledge is “knowledge that can only be performed, such as how to swim, do mental arithmetic, or speak fluently” (DeKeyser, 2017, p. 17). In order to transfer declarative knowledge to procedural knowledge (proceduralization), the former and the deliberate practice of a target behavior are necessary, and the former needs to be kept in mind when practicing the target behavior.
Declarative and procedural knowledge are sometimes used interchangeably with the terms explicit and implicit knowledge, respectively. However, the distinction between explicit and implicit places the emphasis on an awareness of knowledge. That is, “explicit knowledge is knowledge one is aware of, and implicit knowledge is knowledge without awareness” (DeKeyser, 2017, p.16). In this regard, DeKeyser (2009, p. 121) stated that “declarative knowledge is not necessarily explicit because it is not necessarily accessible to awareness” and “procedural knowledge is not necessarily implicit, because it can be the result of proceduralization (and partial) automatization of declarative knowledge, and still allow or even require a certain degree of conscious access when being used.” L2 learners in instructed contexts have limited opportunities for exposure to L2 input, and they may not be able to develop implicit L2 knowledge easily. The participants in this study were L2 learners in instructed contexts, and procedural knowledge developed from declarative knowledge, learned from teachers or textbooks, may be easily acquired than implicit knowledge. Therefore, this study employs the developmental dichotomy of declarative/procedural knowledge.
1.2 Measurements of Declarative/Procedural KnowledgeThe seminal study on measures of implicit and explicit L2 knowledge was conducted by Ellis (2005), although these measures could also apply to assess declarative and procedural knowledge (Zhao & Ellis, 2020). In his study, Ellis developed three types of measures for implicit knowledge: a timed grammaticality judgment test (TGJT), an elicited imitation test (EIT), and an oral narration test (ONT). Further, he developed two types of measures for explicit knowledge: an untimed grammaticality judgment test (UGJT) and a metalinguistic knowledge test (MKT). In devising these measures, he set seven criteria for operationalizing the two types of knowledge: degree of awareness, time available, focus of attention, systematicity, certainty, metalinguistic knowledge, and learnability. For instance, in terms of degree of awareness, Ellis (2005) mentioned that “this criterion refers to the extent to which learners are aware of their own linguistic knowledge” (p. 151). Thus, to elicit the implicit knowledge of L2 learners, they are required to respond to a task intuitively. Conversely, to elicit the explicit knowledge, learners are required to consciously respond to a task using their L2 knowledge. Concerning time availability, Ellis (2005) stated that “this criterion is concerned with whether learners are pressured to perform a task online or whether they have an opportunity to plan their response carefully” (p. 151). Therefore, to elicit the implicit or explicit knowledge of L2 learners, they are required to perform a task with or without a time constraint. Ellis (2009) showed that a Confirmatory Factor Analysis supported their prediction. That is, the EIT, the ONT, and the TGJT were loaded on one factor (implicit knowledge), whereas the UGJT and the MKT were loaded on a second factor (explicit knowledge). Thus, he concluded that these are relatively separate measures of implicit and explicit L2 knowledge. Several other researchers (e.g., Bowles, 2011; Gutiérrez, 2013; Zhang; 2014) have validated Ellis’s five measures. For instance, Zhang (2014) validated Ellis’s test battery by employing four of his tests (EIT, GJT, UGJT, and MKT) for Chinese learners of English. The results showed that, similar to those of Ellis’s study (2005), the learners’ scores were loaded on two factors and distinguished implicit and explicit knowledge.
However, recently, several researchers (e.g., Serafini & Sanz, 2016; Suzuki & DeKeyser 2015; Suzuki & DeKeyser 2017; Vafaee, Suzuki, & Kachinske, 2017) distinguished implicit knowledge and automatized explicit knowledge based on the degree to which awareness was involved during tasks and claimed that Ellis’s implicit knowledge tests measure automatized explicit knowledge. For instance, Suzuki and DeKeyser (2015) argue that advanced L2 learners may use automatized explicit knowledge even under a time constraint condition, such as the EIT. Further, Vafaee et al. (2017) state that the GJT is instead the explicit end of the continuum. Moreover, Serafini and Sanz (2016) employed the EIT as a measure of “explicit knowledge that is more or less automatized depending on the amount and quality of practice” (p. 13) because they were doubtful that late L2 learners in the instructed context can develop implicit knowledge. Thus, for L2 learners in instructed contexts, the EIT and the TGJT could be used as measures of procedural knowledge that is more or less automatized and the UGJT and MKT could be used as measures of declarative knowledge. As the EIT and TGJT would require more automatic processing because of the time constrains, the tests would encourage the use of procedural knowledge. In contrast, as the UGJT and MKT allow learners to access their declarative knowledge because there is no time constraint, they would use more declarative knowledge in the tests. However, as Ellis stated (2005), these tests are not pure measures of explicit (declarative)/implicit (procedural) knowledge.
1.3 Relationships Between Declarative/Procedural Knowledge and L2 ProficiencyDespite the use of the terms “explicit” and “implicit” in place of “declarative” and “procedural”, several studies have examined the relationships between declarative and procedural knowledge and L2 proficiency (Elder & Ellis, 2009; Zhang, 2015). In Elder and Ellis’s studies (2009), Study 1 investigated the relationships between implicit and explicit knowledge and L2 proficiency derived from the computer- and the internet-based TOEFL. Study 2 investigated the relationships between the two types of knowledge and the IELTS. The results showed that all subsections (speaking, listening, reading, and writing) of the TOEFL tests were related to explicit knowledge alone, while the IELTS was related to both implicit and explicit knowledge. Although the measures of implicit knowledge were different between the two studies (Study 1 adopted the TGJT alone, whereas Study 2 employed the EIT and TGJT), they revealed different relationship patterns between the two types of knowledge and the two standardized proficiency tests. In addition, in Study 2, different relationship patterns were revealed between the oral (speaking and listening) and written (writing and reading) sections of the IELTS test. That is, in general, the oral sections were more related to implicit knowledge than explicit knowledge, while the written sections had stronger correlations with explicit knowledge than implicit knowledge.
Further, Zhang (2015) examined the relationships between the two types of knowledge measured by the EIT, TGJT, UGJT (only ungrammatical items), and MKT, and general L2 proficiency assessed by the Oxford Placement Test (OPT). A correlation analysis revealed that the test scores, including both the listening and grammar sections (general proficiency), were significantly correlated with both implicit and explicit knowledge. Although the grammar section also correlated with both implicit and explicit knowledge, the listening section was significantly correlated with implicit knowledge alone. Moreover, three multiple regression analyses showed that 42.2% of the general proficiency score was explained by the EIT and TGJT, 11.4 % of the listening score by the TGJT, and 48.2 % of the grammar score by the EIT, UGJT, and MKT. The beta value suggested that the EIT predicts general and grammar proficiency more effectively than other variables.
1.4 The Present StudyThe results of the previous studies suggest that different proficiency tests may tap different types of knowledge. However, as the number of studies is limited, more empirical research is needed (Elder & Ellis, 2009; Zhang, 2015). Thus, this study examines the relationships between declarative/procedural knowledge and L2 proficiency derived from the TOEIC Bridge test. The TOEIC Bridge test was selected as a measure of L2 proficiency for four reasons. First, previous studies have not used it to investigate the relationships between the two types of knowledge and L2 proficiency. Second, the TOEIC Bridge test is intended for beginner and lower intermediate learners and was suitable for the participants in this study. Third, the TOEIC Bridge is appropriate for examining Elder and Ellis’s results from Study 2 (2009). That is, oral skills are more related to implicit (procedural) knowledge than explicit (declarative) knowledge, while written skills are more related to explicit (declarative) knowledge than implicit (procedural) knowledge. Finally, the TOEIC Bridge test is used in several high schools, technical colleges, and universities, and clarifying the type of knowledge measured by the TOEIC Bridge test could provide useful information for high school, technical college, and university English teachers/researchers. For instance, Yoshida (2009) highlights that the TOEIC Bridge test is one of the major placement tests used in Japan. The research questions (RQs) of this study are as follows:
One hundred forty-six Japanese university students who had been learning English in instructed contexts for about six years participated in this study. Their English proficiency level was upper beginner to lower intermediate. All the participants understood the purpose of the study and signed a consent form.
2.2 MaterialsTo measure the declarative and procedural knowledge of the participants, four of Ellis’s tests (2005, 2009) were employed (see appendices). The EIT and TGJT were used to measure procedural knowledge of the participants, while the UGJT and MKT were used to measure the declarative knowledge of the participants. The four tests included the same 17 English grammatical structures as used in Ellis’s study (2005, 2009). He chose the structures for four reasons. First, they “were known to be universally problematic to L2 learners.” Second, they represented “both early and late acquired grammatical features.” Third, they represented a “broad range of L2 proficiencies.” Finally, they “included both morphological and syntactic features” (Ellis, 2005, p. 154).
The EIT was composed of 34 statements about the participants (17 grammatical and 17 ungrammatical). Each sentence included one of the 17 grammatical structures. The sentences were audio-recorded, and the participants were required to listen to each sentence and write down if they found them to be true or false in their answer sheet. For instance, if they heard the statement “I can speak English very well,” they would have to indicate if it was true in their own case. Subsequently, they reproduced the sentences orally, which was then recorded. The interval between each sentence was 6 seconds. The participants’ attention was first directed to the meaning, rather than the form, as they were required to answer questions about themselves before reproducing them. This prevented the rote repetition of the sentences. However, Ellis’s original sentences were relatively difficult, and the contents of the statements were not suitable for the Japanese participants of this study. Therefore, the sentences were adjusted to accommodate their level, although the 17 grammar structures were the same as those used in Ellis’s study (2009).
The contents of the TGJT and UGJT were identical. The two tests comprised 68 sentences (34 grammatical and 34 ungrammatical) and each sentence included one of the 17 grammar structures. The sentences were presented on a PowerPoint slide show. The TGJT required participants to judge whether the sentence was grammatical or ungrammatical within 3.5 seconds. The imposition of this time limit was followed by Erçetin and Alptekin (2013). Conversely, there was no time constraint for the UGJT. The two tests required the participants to judge the sentence on a computer screen and write their responses on their answer sheet. For the TGJT, after 3.5 seconds, the sentence on the screen was automatically replaced by the next sentence. However, in the UGJT, the participants were required to press the “Enter” key to proceed to the next sentence.
The MKT test comprised two parts. Part 1 required participants to read 17 ungrammatical sentences, which included one of the 17 grammatical structures, and they were told to choose the answer from the four options provided that best explained the sentence error. Part 2 had two subsections. Section 1 required participants to find examples of 21 grammatical terms in a given passage. Section 2 required them to find the named grammatical item in a sentence. The MKT was identical to the one used in Ellis’s study (2009) except that all options that explained the error in the sentence in Part 1 and the grammatical terms in Part 2 were translated from English (L2) into Japanese (L1) as some of the L2 terminology used in the test was unfamiliar to the participants. The MKT did not impose any time constraint on the participants, and they wrote their responses on their answer sheet.
Time constraints in the EIT and TGJT encourage participants to use their procedural knowledge as they force relatively automatic processing of L2. Although Ellis (2005) considered the tests as measures of implicit knowledge, several researchers (Serafini & Sanz, 2016; Suzuki & DeKeyser, 2015) argued that they were measures of automatized explicit knowledge, as the learners can still access their explicit knowledge on the tests. Considering the fact that the participants in this study were L2 learners in instructed contexts and had not had a rich exposure to the L2, what they were likely to use was primarily procedural knowledge that developed from declarative knowledge and is more or less automatized. On the other hand, in the UGJT and MKT, participants can access their declarative knowledge, as there is no time constraint. Thus, these tests would encourage them to use more declarative knowledge. Although these tests are not pure measures of declarative or procedural knowledge, for this study, they were used as measures that tap into more of their declarative or procedural knowledge.
The TOEIC Bridge test was used as the L2 proficiency measure in this study. The TOEIC Bridge test is composed of listening and reading sections and comprises 100 multiple-choice questions. The test takes approximately 60 minutes to complete. The scores on the listening and reading sections range from 10 to 90 and the total score ranges from 20 to 180.
2.3 ProcedureTo measure the participants’ L2 proficiency, the TOEIC Bridge test was completed one week before the four tests were administered. The participants took the two procedural knowledge tests (the EIT and TGJT) individually and the two declarative knowledge tests (the UGJT and MKT) in groups. They took the tests in the same order as that followed by Ellis’s participants (2005): the EIT, the TGJT, the UGJT, and finally the MKT. All tests were conducted using a computer.
2.4 ScoringRegarding the EIT, the participants’ reproduction of sentences was scored based on the three criteria used in Ellis’s study (2005). That is, if a participant’s response included the target structure, it was given 1 point. However, if the target structure was not included or incorrectly included in their response, it was considered incorrect and given 0 points. Regarding the TGJT, the UGJT, and the MKT, if the correct answer was chosen, 1 point was provided while no points were provided when an incorrect answer was selected.
2.5 AnalysisDescriptive statistics of the five measures were obtained. Cronbach α for declarative and procedural knowledge measures were calculated to confirm their internal consistency reliability. Pearson product-moment correlations were computed to investigate the relationships between the measures.
First, internal consistency was computed using Cronbach’s α. The reliability scores of the EIT, the TGJT, the UGJT, and the MKT tests were α = .82, α = .56, α = .69, α = .84, respectively.
Table 1 presents the descriptive statistics of the EIT, TGJT, UGJT, MKT, and TOEIC Bridge scores (total, listening and reading). To examine RQ1 and RQ2, Pearson product-moment correlations were run. The results are shown in Table 2. The general L2 proficiency (Listening + Reading) measured by the TOEIC Bridge test was significantly and moderately correlated to all declarative and procedural knowledge measures. The correlation between the EIT, which was a procedural knowledge test, and general L2 proficiency (r = .622, p < .01) was the strongest among the correlations between general L2 proficiency and other declarative and procedural knowledge measures. However, the other procedural knowledge measure, TGJT, had a weaker correlation with general L2 proficiency (r = .437, p < .01) than the declarative knowledge measures (r = .553, p < .01). Further, the subsections of the TOEIC Bridge were significantly correlated with all declarative and procedural knowledge tests. Among the declarative and procedural knowledge tests, the listening section had the strongest correlations with the EIT, which was a procedural knowledge test (r = .586, p < .01). However, the listening proficiency had the second-strongest correlations with the MKT test, which was a declarative knowledge test (r = .491, p < .01). Moreover, reading proficiency had a stronger association with the EIT (r = .582, p < .01) than any other declarative or procedural test. However, it had the second-strongest association with the UGJT (r = .539, p < .01). Regarding the correlations between the declarative and procedural knowledge tests, the EIT had significant but weak correlations with the TGJT (r = .244, p < .01), the UGJT (r = .313, p < .01) and the MKT (r = .352, p < .01). The TGJT had significant and moderate correlations with the UGJT (r = .608, p < .01) and MKT (r = .448, p < .01). The UGJT had a significant and moderate association with the MKT (r = .566, p < .01) in addition to the other correlations reported above.
| Tests | N | Minimum | Maximum | M | SD |
| Bridge (L+R) | 146 | 84 | 170 | 122.0 | 18.3 |
| Listening | 146 | 44 | 84 | 60.5 | 7.9 |
| Reading | 146 | 38 | 86 | 61.6 | 11.6 |
| EIT | 146 | 0 | 23 | 11.6 | 5.1 |
| TGJT | 146 | 24 | 57 | 41.9 | 5.7 |
| UGJT | 146 | 19 | 62 | 45.1 | 6.5 |
| MKT | 146 | 1 | 39 | 24.4 | 6.9 |
Note. Bridge = TOEIC Bridge test; L = Listening; R = Reading; EIT = elicited imitation test; TGJT = timed grammaticality judgment test; UGJT = untimed grammaticality judgment test; MKT = metalinguistic knowledge test.
| Bridge (L+R) | Listening | Reading | EIT | TGJT | UGJT | MKT | |
| Bridge (L+R) | 1 | .909** | .959** | .622** | .437** | .553** | .553** |
| Listening | 1 | .754** | .586** | .399** | .488** | .491** | |
| Reading | 1 | .582** | .419** | .539** | .538** | ||
| EIT | 1 | .244** | .313** | .352** | |||
| TGJT | 1 | .608** | .448** | ||||
| UGJT | 1 | .566** | |||||
| MKT | 1 |
Note. Bridge = TOEIC Bridge test; L = listening; R = reading; EIT = elicited imitation test; TGJT = timed grammaticality judgment test; UGJT = untimed grammaticality judgment test; MKT = metalinguistic knowledge test.
*p < .05, **p < .01 (two-tailed tests)
In terms of RQs 1 and 2, the correlation analyses showed that both declarative and procedural knowledge were significantly related to general L2 proficiency, listening, and reading. The EIT, which was a measure of procedural knowledge had the strongest correlations with the L2 proficiency measures. Thus, the TOEIC Bridge test may encourage more use of procedural knowledge than declarative knowledge. The TOEIC Bridge test required the participants to answer 100 questions in one hour. This relatively short time-limit may have encouraged them to use their procedural knowledge, leading to the positive relation between procedural knowledge and the test. However, as moderate correlations between declarative knowledge and L2 proficiency measures were also found, the test may also have tapped into the learners’ declarative knowledge. Thus, as the IIBC (2019) insists, the test would be appropriate for beginners and lower intermediate learners. That is, the beginners can rely on the initial declarative knowledge, while the lower intermediate learners can rely on more developed procedural knowledge.
However, TGJT, the other procedural knowledge measure, had weakest correlations with L2 proficiency tests among the declarative and procedural knowledge tests. There are several possible reasons for this contradictory result. First, as Vafaee et al. (2017) insisted, the TGJT test might have encouraged the use of declarative or explicit knowledge rather than that of procedural or implicit knowledge. Actually, the correlations between the TGJT and the declarative knowledge measures were stronger than the correlation between the TGJT and EIT, the other procedural knowledge measure. Another possible reason for the contradiction might have derived from the relatively low reliability of the TGJT test in this study. This study included beginner-level L2 learners. For those learners, the time constrained GJT may have been too difficult, and they may have been unable to read all parts of each sentence presented. That is, they may have been able to read short sentences but unable to finish reading long sentences within the time limit, which may have led to a relatively low reliability. Thus, their answers may have reflected only guesses.
Elder and Ellis (2009) insisted that implicit knowledge is more related to oral skills whereas explicit knowledge is more connected with written skills, and their argument may be partly applicable to the results of this study. The results of this study showed the slightly stronger relationship between the EIT and listening than between the EIT and reading. Also, they demonstrated the stronger relationships between the UGJT or MKT and reading than the UGJT or MKT and listening. Thus, the results suggest that listening may be more related to procedural knowledge, while reading may be more related to declarative knowledge. Compared to the listening section, the participants would have been able to access more of their declarative knowledge in the reading section as they could sometimes reread and analyze sentences. Therefore, the declarative knowledge may have been more related to reading (written skills) than listening (oral skills).
Elder and Ellis (2009) showed that TOEFL was related to the UGJT and the MKT (declarative knowledge measure), while IELTS were associated with the EIT/TGJT (procedural knowledge measures) and UGJT/ MKT (declarative knowledge measures). The results found in this study may be more similar to those of the IELTS. That is, both declarative and procedural knowledge may be related to L2 proficiency. Although Elder and Ellis did not find a significant relationship between procedural knowledge and TOEFL, they did not include the EIT in their study. Moreover, TGJT was the only measure used for procedural knowledge and it may not have encouraged the use of procedural knowledge, as the results of this study suggest. This may contribute to the difference between the results of the TOEIC Bridge or IELTS and TOEFL.
The limitation of this study is that the TOEIC Bridge test did not include assessment of production skills at the time this study was carried out, and this study only measured the participants’ receptive skills. Thus, the relationships between production skills and declarative/procedural knowledge were not revealed; further studies must investigate this. Moreover, as the TGJT used here as one of the procedural knowledge measures was more correlated with the declarative knowledge measures than with the procedural knowledge measure, it might have measured participants’ declarative knowledge rather than their procedural knowledge. Thus, another measure that taps into procedural knowledge should be incorporated into future studies to reveal the relationship between procedural knowledge and L2 proficiency tests more clearly. Furthermore, as this study did not include advanced level learners, who are assumed to possess more procedural knowledge, the relationships between declarative/procedural knowledge and the TOEIC Bridge test were not examined for these learners. It is expected that the correlations between procedural knowledge and L2 proficiency measures might be stronger, as they would use more procedural knowledge for solving test questions.
This study investigated the relationships between L2 proficiency assessed by the TOEIC Bridge test and declarative/procedural knowledge. The results showed that the L2 proficiency was related to both procedural and declarative knowledge. However, procedural knowledge was more associated with L2 proficiency, likely because the TOEIC Bridge asked the learners to answer questions in a relatively short time and encouraged the use of procedural knowledge. In addition, the results of this study suggest that listening skills might be more related to procedural knowledge than reading skills, whereas reading skills might be more related to declarative knowledge than listening skills. However, these relations might change if another L2 proficiency test is used. The use of tests for more advanced learners such as TOEIC Listening & Reading tests or EIKEN tests (Grade Pre-1 and Grade 2) would be more promising as these tests are popular in Japan. As pedagogical implications, this study suggests that L2 teachers should choose or create an L2 proficiency test carefully, considering not only their students’ L2 proficiency level but also the type of knowledge they want to measure, whether it is declarative knowledge, procedural knowledge, or both. For instance, if they teach their students the L2 in a form-oriented way and need to assess their declarative knowledge, procedural knowledge-oriented tests like the TOEIC Bridge test might not be the best option, although the correlations between the test and declarative knowledge were still found.
The results of this study suggest that different L2 proficiency tests may measure varying degrees of declarative/procedural knowledge in L2 learners. However, further studies are needed to investigate the relationships between other L2 proficiency tests and declarative/procedural knowledge to reveal the constituents of L2 proficiency.
Regular past tense
I watched a soccer game yesterday.
*I watch a baseball game yesterday.
Since and for
I have been living in Japan for 3 years.
*I have been living in Otawara since 3 years.
Appendix B: Samples of the TGJT/UGJT (adapted from Ellis, 2009)*He wants buying a car
He plays volleyball very well.
*Did Keiko finished her homework?
Appendix C: Samples of the MKT (adapted from Ellis, 2009)Part1
下線部の間違いを説明しているものとして適切なものを選びなさい。
Keiko grew some rose in her garden.
a. ‘rose’は数えられるので‘some’の後は複数形である
b. ‘rose’の前に誤った形容詞が使用されている
c. 名詞の前にはaかtheが必要である
d. 可算名詞の前にはsomeではなくa fewである
Part2
適切な語を抜き出しなさい。
Section 1
The materials are delivered to the factory… (動詞、名詞・・・)
Section 2
Joe had nowhere to stay. (不定詞)
