Idiomatic Knowledge and Attitudes toward Idiom Learning: A Case of Kuwaiti EFL College Students

The study attempts to determine if idiomatic knowledge has an influence on students’ attitudes toward learning English idioms. Primarily, the study seeks to uncover if idiomatic knowledge affects Kuwaiti EFL undergraduate college students’ attitudes toward (1) the importance of idiom learning, (2) the difficulties of idiom learning and (3) learning strategies for idioms. Results showed that students had positive attitudes toward English idiom learning. Significant differences in the results were obtained for students’ knowledge of idioms.


Difficulties of Idiom Learning
Idioms are considered a stumbling block for ESL/EFL learners (Andreou & Galantomos, 2007;Cooper, 1999;Irujo, 1984Irujo, , 1986bLiontas, 2002;Nippold, 2003). There are several factors which make idioms difficult to learn in a second or foreign language context. According to Irujo (1986a), ESL/EFL learners struggle with proper usage of idioms. Learners find idioms hard to use correctly in proper context. Researchers explored whether ESL/EFL learners avoid idioms (Irujo, 1993;Laufer, 2000;Yorio, 1989). Avoidance indicates a difficulty that learners may have with the avoided form. Avoidance presumes some knowledge of the avoided form, as one cannot avoid something that he/she does not know. Yorio (1989) investigated the use of idioms by ESL college students and compared it with that of native speakers of English. He analyzed written production of idioms in both groups and found that native speakers used more idiomatic expressions than L2 learners. Moreover, he found that L2 learners not only used fewer idioms in their writing, but also used idioms incorrectly. By contrast, Laufer (2000) showed that idioms were not avoided as a category but some specific idiom types were avoided. She attributed avoidance of specific idiom types to L1-L2 degrees of similarity.
Several studies have focused on the role of L1 in L2 idiom comprehension and concluded that L1 affects L2 idiom comprehension; the effects, however, are both negative and positive (Irujo 9886a;Liontas 2001Liontas , 2002bLiontas , and 2003. Irujo (1986a) conducted one of the earliest studies investigating the influence of learners' L1 on L2 idiom comprehension. Her participants were Venezuelan advanced learners of L2 English, who were attending college in the United states. In her study, the idioms used were divided into three categories based on similarity between L1 and L2: identical idioms, meaning those that are identical in form and meaning in L1 and L2; similar idioms, which have similar meaning and form in L1; and different idioms, which have no similarity to L1. The data revealed that transfer from L1 is affected by the degree of similarity of the idioms between the two languages. Identical idioms were the easiest to comprehend and produce, followed by similar idioms, which showed interference from the native language. Different idioms were the most difficult to comprehend and produce, but showed less interference from the native language. Similar results were obtained in three studies by Liontas (2001Liontas ( , 2002bLiontas ( , and 2003. Another reason for the difficulty of learning idioms is related to their nonliteral meaning. Idioms are difficult to master because their meanings are unpredictable. As Cooper (1999) elucidates, "because figurative meaning is unpredictable, idioms present a special learning problem for virtually all groups of learners" (p. 233). Irujo (1986b) believes that idioms are problematic because they have nonliteral meanings, but can also have literal counterparts that might confuse learners. The degree of agreement between the literal and nonliteral meanings of idioms is called transparency (Cain et al. 2005). The higher the transparency, the easier it is to derive the meaning of the idiom. The least transparent are described as opaque. Hence, transparent idioms are easier to comprehend than opaque idioms (Cooper 1998;Gibbs 1993). Additionally, idiom familiarity is the extent to which people are said to be familiar with the meaning of the idiom (Abel, 2003;Hubers et al. 2019;Nordmann et al. 2014). Studies indicated that most familiar idioms are easier to understand than less familiar ones (Cronk & Schweigert, 1992Titone & Connine, 1994).
Furthermore, idioms incorporate aspects of the larger cultural context in which the target language arose (Bachman, 1990). One of the main reasons that makes idioms difficult to interpret is their cultural specificity. Researchers studied the influence of etymological elaboration (Boers et al. 2007(Boers et al. , 2008(Boers et al. , 2009) and familiarity with culture-bound expressions (Boers et al. 2001(Boers et al. , 2004 as ways to foster figurative language in L2. Idioms can be a source of potential misunderstanding when culture-specific knowledge is implied. Sometimes idioms reflect themes and ideas that are absent in the learner's own culture, which would hinder comprehension. Elkilic (2008) pointed out that idioms are culture-bound, and thus it is not easy to understand their meaning even if they are transparent. He further (1990) stated that "knowledge about idiomacity is not based on linguistic analysis, but it is essentially a matter of cultural awareness." (p. 31). For example, the idiom carry coals to Newcastle in British culture means to take something to a place where it is not needed. Since cross-cultural variation can have an impact on the learner's interpretation of idioms, Boer and Demecheleer (2001) suggested that the best approach to teaching idioms is to benefit from the teacher's awareness of cross-cultural differences.

Comprehension Strategies for L2 Idioms
The use of contextual information is one of the major strategies employed by L2 learners to comprehend idioms. Cooper (1999) identified eight major strategies that L2 learners employed when analyzing idioms: guessing from context, using literal meaning, referring to L1, and using other strategies. Guessing from context was the most frequent employed strategy, used in 28% of instances; using the literal meaning was also among the frequent strategies (19%). However, there were individual variations across participants and within participants. Cooper (1999) explained these variations by referring to comprehension and that it is a dynamic process. Along the same lines, Bulut and Celik-Yazici (2004) conducted a study investigating idiom-processing strategies of eighteen Turkish teachers of English. The participants used a heuristic approach when they encountered unknown idioms, with guessing from context ranked as the most -used strategy. This led the researchers to hypothesize that L2 learners will first resort to contextual information to interpret idioms. In line with Cooper (1999), they also found that L2 learners do not use a single strategy but a variety of strategies as appropriate. Similarly, studies showed that learners rely on context and L1 to decode idioms (Liu, 2008;Saleh & Zakaria, 2013). Other studies demonstrated that learners utilize keywords (Guduru, 2012) and mnemonic strategies like etymological elaboration (Boers et al., 2004).

Attitudes and Knowledge of Idioms
Studies on attitudes of Kuwaiti Arabic learners of English and idiom learning are exceptionally rare (Al-Houti & Aldaihani, 2018). Al-Houti and Aldaihani (2018) investigated Kuwaiti EFL students' attitudes toward learning English idioms; the researchers examined two learner-related variables (age and year of study) and concluded that Kuwaiti EFL college students had positive attitudes toward idiom learning overall, but that the results indicated significant differences in attitudes toward English idiom for the independent variable age. Research had shown that many EFL students are handicapped in their ability to comprehend idioms and have poor idiomatic competence (Adkins, 1968;Hussein, Khanij & Makhzoomy, 2001;Katsarou, 2011). A very small number of studies have been devoted to Arabic EFL/ESL learners and idiomatic knowledge (Aljabri, 2013;Al-Khawaldeh et al., 2016;Al-Kadi, 2015). Al-Houti and Aldaihani (2020) conducted the sole existing study that tested Kuwaiti EFL learners' knowledge of frequently used English idioms. The results showed that these learners have difficulty comprehending common idiomatic expressions. The relationship between idiomatic competence of EFL learners and their perceptions of idiom learning is seldom investigated in the literature. Tran (2013) explored this relationship by examining university students in Vietnam; the results revealed that students had poor idiomatic competence while, more interestingly, uncovering a paradox between learners' idiomatic knowledge and their desire to learn idioms. Due to a paucity of research on the subject, the purpose of this study is to examine the effect of idiomatic knowledge on students' attitudes toward idiom learning. The research questions of the present study are: 1. Do EFL college students' attitudes toward idiom learning differ based on their preexisting knowledge of idioms? 2. Do EFL college students with different levels of idiomatic knowledge differ in their attitudes toward the importance of idioms, the difficulties of idioms, and idiom-learning strategies? 3. Do EFL college students with different levels of idiomatic knowledge differ in their attitudes toward the individual survey items?

Participants
A total of 218 female EFL undergraduate students participated in the study. The participants were English major students at the CBE (College of Basic Education) in Kuwait. All of the participants who took part in the study were native speakers of Arabic. Their age ranged between 18 and 41 years old. They were chosen randomly to participate in the study.

Instruments
An idiomatic knowledge test, consisting of 27 frequently used idioms, was designed to assess the students' knowledge of frequently used idioms. The selection of idioms was based on several criteria: (1) idioms from among the 300 most common English idioms, according to the frequency analysis of Collins COBUILD data, (2) idioms that can be used in both British and American English, (3) idioms covering various structural patterns, (4) idioms covering the spectrum of idiomaticity, and (5) idioms that have equivalents in Arabic (identical or similar) as well as those that lack Arabic equivalents. The designed test consisted of matching, multiple-choice, and gapfill tasks. The matching task requires the participants to match the meanings and the idioms. The multiple-choice task requires selecting the most appropriate idiom for each situation. The gap-fill task requires completing the idioms based on meaning.
To measure students' attitudes toward learning English idioms, the researchers created a questionnaire to fulfil this purpose after an exhaustive review of relevant studies (see, e.g., Cooper, 1999;Guduru, 2012;Irujo 1986;Khan & Can Daşkin, 2014;Liu, 2008;Pimenova, 2011;Saleh & Zakaria, 2013;Tran, 2013). The questionnaire consisted of 19 items with a five-point Likert scale (1 = strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, and 5 = strongly agree). The 19 items consisted of three subscales of students' attitudes toward learning English idioms: (1) attitudes toward importance of English idiom learning, (2) attitudes toward difficulties of English idiom learning, and (3) attitudes toward learning strategies of idioms. Table 1 presents the questionnaire items and their corresponding subscales. English language teachers should teach students idioms. Q4 Idioms should be included in language curricula. Q5 Idioms are important because they are frequent and common. Q6 Idioms are important because they can communicate aspects in a nonliteral (figurative) way. Q7 Idioms are important because they reflect and represent culture. Q8 Idioms are important because they indicate native-like knowledge of the language as they reflect proficiency.

Procedure
To establish the validity of the idiomatic knowledge test, specialists in the English language and literature department were asked to review it. They judged the selected idioms as highly familiar. A pilot study was conducted on 35 female EFL undergraduate students (who were excluded from the main study) to verify the reliability and validity of the instrument. The researchers administered the test to measure the students' idiomatic knowledge. The students took the test in a quiet classroom at CBE. Students were asked to complete the test at their own pace. A Cronbach's alpha reliability coefficient of 0.77 was calculated, suggesting that reliability of the instrument was satisfactory.
As to the questionnaire, it was also piloted with a small group of EFL undergraduate students to check the clarity of the items before the actual implementation, as well as to verify the reliability and validity of the instrument. As a result, some modifications were made to improve the questionnaire. The revised version of the Journal of Education and Practice www.iiste.org ISSN 2222-1735 (Paper) ISSN 2222-288X (Online) Vol.12, No.27, 2021 questionnaire was distributed along with the idiomatic knowledge test. Hence, participants who took the idiomatic knowledge test also completed the questionnaire. A Cronbach's alpha was above .70, suggesting that reliability of the instrument was satisfactory. All subjects gave informed consent to participate in the study, and ethical approval was obtained.

Data Analysis
The questionnaire items contained three subscales of students' attitudes toward learning English idioms. For each subscale, a composite score was created by averaging the responses of the corresponding survey items. The composite scores ranged from 1 to 5, with higher scores indicating more positive attitudes toward the subscale. Normality of the data was assessed via the Shapiro-Wilk test. A p-value < 0.05 on the Shapiro-Wilk test indicated that data were not normally distributed. Kruskal-Wallis tests (Hollander and Wolfe, 1999) were used to determine if there was a statistically significant difference in the 1) three subscales of EFL college students' attitudes toward learning English idioms, and 2) the responses to each individual survey item among participants with different knowledge of idioms. If the results of Kruskal-Wallis test were significant, Dunn's procedure (Dunn, 1964) for pairwise comparisons was performed. For the tests, p-values less than 0.05 indicated significance. All analyses were conducted using SPSS version 23 (IBM Corp., Armonk, NY).

Results and Discussion
The research question of the present study asked: Do EFL college students' attitudes differ based on their knowledge of idioms? To answer this research question, Kruskal-Wallis tests were used to determine if there was a statistically significant difference in the 1) three subscales of EFL college students' attitudes toward learning English idioms, and 2) the responses to each individual survey item among participants with different knowledge of idioms.

Results of the three subscales according to knowledge of idioms
The results of the Kruskal-Wallis tests for the three subscales are presented in Table 2.

Results of the Kruskal-Wallis Tests for the Three Subscales
Knowledge of idioms 0-6 (N = 32) 7-13 (N = 129) There was a statistically significant difference in attitudes toward difficulties of English idiom learning among participants with different knowledge of idioms (p = 0.042). In particular, the results of the pairwise comparisons indicated that people with the lowest marks (0-6) had statistically significantly more positive attitudes toward the difficulties of English idiom-learning than people with the highest marks (14-27) (M = 3.51 for 0-6, M = 3.12 for 14-27, p = 0.037); there was no statistically significant difference in attitudes toward the difficulties of English idiom-learning between knowledge of idioms 0-6 and 7-13 (p = 0.330), and between knowledge of idioms 7-13 and 14-27 (p = 0.410).
There was a statistically significant difference in attitudes toward learning strategies for idioms among participants with different knowledge of idioms (p = 0.031). However, the results of the pairwise comparisons indicated that there was no statistically significant difference in attitudes toward learning strategies for idioms among participants with different knowledge of idioms (p = 0.055 for 14-27 vs. 7-13; p = 0.084 for 14-27 vs. 0-6; p = 1.000 for 7-13 vs. 0-6). Thus, we concluded that there was no statistically significant difference in attitudes toward learning strategies for idioms among participants with different knowledge of idioms.
There was no statistically significant difference in attitudes toward importance of idioms among participants with different knowledge of idioms (p = 0.050).

Results of the individual survey items according to knowledge of idioms
The results of the Kruskal-Wallis tests for the individual survey items are presented in Table 3.

Results of the Kruskal-Wallis tests for the Survey Items
Knowledge of Idioms 0-6 (N = 32)   Table 3, it is apparent that students with the highest marks (14-27) recognized the importance of idioms more than students with lower marks (7-13 and 0-6), as those students have more positive attitudes toward all of the survey items related to the importance of idioms (Q1 to Q8). Specifically, those students realized more than students with lower knowledge of idioms that idioms are important for learners and that they should be included in language curricula. Also, they agreed more than students with lower knowledge of idioms that it is necessary for English language learners to have knowledge about idioms. Furthermore, they had more positive attitudes toward the importance of idioms and its link to frequency, figurativeness, and culture than students with lower knowledge of idioms. However, the majority of these observed differences in the Journal of Education and Practice www.iiste.org ISSN 2222-1735 (Paper) ISSN 2222-288X (Online) Vol.12, No.27, 2021 responses across participants with different knowledge of idioms were not significant (p > 0.05). There was a statistically significant difference in the responses of Q2 (English language learners should have knowledge about idioms) among participants with different knowledge of idioms (p = 0.018). In particular, the results of the pairwise comparisons indicated that students with the lowest marks (0-6) had agreed statistically significantly less to Q2 than students with the highest marks (14-27) (M = 4.25 for 0-6, M = 4.28 for 14-27, p = 0.022); there was no statistically significant difference in responses of Q2 between knowledge of idioms 0-6 and 7-13 (p = 0.596), and between knowledge of idioms 7-13 and 14-27 (p = 0.098).
As indicated in Table 3, students with the highest marks (14-27) agreed less than students with lower marks (7-13 and 0-6) with regard to all of the survey items related to the difficulties of idioms (Q9 to Q14). Students with lowest marks (0-6) agreed more than students with better knowledge of idioms that idioms are difficult. They recognized that idioms are challenging. They attributed the difficulty of idioms to frequency and familiarity as well as other factors like figurative meaning, lack of L1 equivalent, and the need for cultural and historical knowledge. Although students with the lowest marks (0-6) agreed more than students with better knowledge of idioms on these sources of difficulties, the difference in the responses was not significantly significant (p > 0.05). There was a statistically significant difference in the responses of Q12 (Idioms are difficult because they are hard to use in contexts) among participants with different knowledge of idioms (p = 0.002). In particular, the results of the pairwise comparisons indicated that students with the highest marks (14-27) had agreed statistically significantly less to Q12 than students with the lowest marks (0-6) (M = 3.34 for 0-6, M = 2.65 for 14-27, p = 0.008) and people with the middle marks (7-13) (M = 3.16 for 7-13, M = 2.65 for 14-27, p = 0.007); there was no statistically significant difference in responses of Q12 between knowledge of idioms 0-6 and 7-13 (p = 1.000).
As a strategy, students with different knowledge of idioms agreed almost equally about the use of contextualization to learn idioms. Students with low marks preferred to use key words to learn idioms whereas students with high marks preferred to learn idioms through context. These responses; however, did not yield a significant difference (p > 0.05). There was a statistically significant difference in the responses to Q15 (I rely on L1 to learn idioms i.e. through L1 translation) among participants with different knowledge of idioms (p = 0.002). In particular, the results of the pairwise comparisons indicated that people with the highest marks (14-27) had agreed statistically significantly less to Q15 than people with the lowest marks (0-6) (M = 3.72 for 0-6, M = 2.61 for 14-27, p = 0.000) and people with the middle marks (7-13) (M = 3.16 for 7-13, M = 2.61 for 14-27, p = 0.015); there was no statistically significant difference in responses of Q12 between knowledge of idioms 0-6 and 7-13 (p = 0.053). As a learning strategy, L1 translation is less preferred by students with good knowledge of idioms than for students with low knowledge of idioms.

Limitations
The primary limitation of the present study is that it was a single-gender study. Hence, generalizations to the opposite gender are limited. This, coupled with the use of one local college, constitutes another major limitation. Data from more than one local college and from both genders will result in a better representation.

Conclusion and Recommendations
The study investigated whether there is an influence of preexisting idiomatic knowledge on EFL college students' attitudes toward (1) the importance of idiom learning, (2) the difficulties of idiom learning, and (3) strategies for learning idioms. From the present study, we conclude that participants had positive attitudes for the subscales. In particular, participants who achieved the highest marks recognized the importance of idioms more than participants with low marks. Additionally, participants who achieved the lowest marks in the idiomatic knowledge test had statistically more positive attitudes toward the difficulties of idiom learning. Therefore, they are aware that idioms are difficult to learn. They attributed the difficulty of idioms to frequency and familiarity, as well as factors like figurative meaning, lack of L1 equivalent, and cultural and historical knowledge. As a learning strategy, L1 translation was preferred by participants with low knowledge of idioms. Also, they preferred to use keywords to learn idioms, whereas participants with good knowledge of idioms preferred to learn idioms through context.
Based on the findings of the current study, EFL instructors are recommended to use L1 translation and key words to teach students idioms if their students do not have good preexisting knowledge of idioms. However, they are recommended to use contextualization with students who already have good knowledge of idioms. Furthermore, EFL instructors should make students aware of any potential negative transfer as a result of the use of L1.