Generally speaking, there are two types of approaches to studying language acquisition: longitudinal and cross-sectional studies (Larsen-Freeman and Long 1991).
Both longitudinal studies and cross-sectional studies have their merits and drawbacks.
Regarding longitudinal studies, the researcher usually investigates a group of participants or an individual case for a long period of time. The benefits are that it is straightforward to examine the natural occurring language and the naturalistic data offer a thorough description of subjects‟ language behavior. However, the drawback may include the limitation of the subject pool and inconvenience for data collection.
Explicitly, this kind of study consists of a small subject pool in consequence of the long-time investigation and the researcher should be as non-intrusive as possible during the whole process of investigation. Owing to the small subject pool, the data obtained from longitudinal studies may not be generalized to other children (Larsen-Freeman and Long 1991).
Compared with the naturalistic context in longitudinal studies, cross-sectional studies provide a controlled context (i.e., the use of an artificial task) for subjects
(Larsen-Freeman and Long 1991). In addition, cross-sectional studies set up several factors beforehand in which one is needed to be constant. Therefore, the advantages are that it would be more efficient to look into extensive issues at one time with a large sample size. In addition, the researcher can generalize the results to the large
46
population instead of merely applying to specific subjects (Larsen-Freeman and Long 1991).
Since the present study aimed to investigate various issues such as age, animacy, task effect, degree of transparency, and metaphoricality. These issues would not be addressed systematically, if the study is conducted longitudinally. Owing to this inadequacy, the present study adopted the cross-sectional means as the research method.
The cross-sectional studies have been generally conducted to explore children‟s ability to comprehend idiomatic or figurative expressions (Levorato and Cacciari 1995, 1999, 2002, 2004, Nippold and Taylor 2002, Cain and colleagues 2005, 2009, Hsieh and Hsu 2010, Vulchanova and colleagues 2011). For example, Levorato and Cacciari conducted a number of cross-sectional studies (1995, 1999, 2002, 2004), which discussed issues in children‟s reading comprehension and semantic analyzability of idioms. Hsieh and Hsu (2010) also employed a cross-sectional approach to investigate effects of familiarity, context, and linguistic convention on L1 idiom comprehension of Mandarin-speaking children.
Previous studies have employed a production task to examine children‟s abilities on figurative expressions (Levorato and Cacciari 1995, 2002). In these studies, researchers coded the collected data into either literal or figurative. However, the purpose of the production task still focused on exploring children‟s comprehension of figurative expressions. The real production of figurative expressions cannot be elicited. In the present study, it would also be difficult to conduct a production task.
For example, if the picture were presented to elicit the production figurative expressions, it would elicit the naming such as laojingcha „an old police officer‟ or laoyisheng „an old man‟; but for items like laopengyou „a friend that one has made for
47
a long time‟ or laohuli „a sophisticated person,‟ it would be difficult to be presented by pictures. Moreover, previous studies showed that children have not started producing conventionalized figurative expressions until the age around 9 (Winner 1988). Since the youngest participants in present study were 7-year-olds, it would be too challenging for them to produce figurative expressions.
The present study investigated children‟s ability to comprehend the figurative expression of lao cross-sectionally. It is hypothesized that children‟s ability correlates possibly with several factors such as age, task effect, animacy, degree of transparency, and metaphoricality. In order to investigate our subjects‟ ability of metaphorical language, we conducted two comprehension tasks in the study: the Interpretation Task (IT Task) and the Multiple-choice Task (MC Task).
The interpretation task consists of questions without alternative choices, whereas the multiple-choice task is composed of questions with three alternative choices. It has been found that multiple-choice questions have been commonly used to examine the comprehension of figurative language because they are more easily presentable and appropriate to assess the understandings of figurative language in context (Levorato and Cacciari 1995, Cain and colleagues 2009).
In the present study, the classification and distribution of the test items are shown in Table 3-2. For more details, please refer to Appendixes A and B.
48 Table 3-2 A Classification of Lao
Type Example IT Task MC Task meanings: Type 1 refers to the core meaning of lao, „old.‟ Type 2 of lao is interpreted non-literally, which can be further divided into two subtypes. Type 2-1 is the transparent non-literal meaning of lao, referring to someone or something is
„long-existing‟. Type 2-1-1 refers to lao, which co-occurs with an animate noun like lao pengyou denoting the long-existing relationship with a friend. Likewise, Type 2-1-2 refers to lao, which appears with an inanimate noun like lao difang „a long-existing place‟. Type 2-2 refers to someone who is „experienced,‟ which is different from Type 2-1; it carries a feature of [+opaque ] representing the relationship
49
with the core meaning of lao is not apparently close and semantically opaque.
Similarly, Type 2-2-1 refers to the opaque non-literal meaning of lao with an animate noun as in lao niao „an expert,‟ and Type 2-2-2 lao appears with an inanimate noun as in lao youtiao „a sophisticated person.‟
In both tasks, there were 15 phrases of lao. The selected items were chosen based on the literality of meanings, either literal or non-literal. Furthermore, non-literal meanings were further divided according to the degree of transparency, either transparent or opaque. For the non-literal meanings with transparent and opaque features, it was separated based on the animacy of the lao NP. Three items were designed for each subtype and five fillers were added. Twenty questions were included in each task. In the interpretation task, the test question was presented in isolation without any alternative answers. As for the multiple-choice task, the test items were adopted from the interpretation task and were presented with three alternative answers.
Previous studies have already claimed that linguistic context is a critical factor affecting the comprehension of figurative expressions (Levorato and Cacciari 1995, 1999, Hsieh 2008, Hsieh and Hsu 2010). Therefore, in the present study two comprehension tasks were presented in the same designed story scenario. There were two main characters in the story: Brother Crab and Little Monkey. The story was described as follows: Brother Crab has been living under the sea for years. One day, he decided to go on an adventure to the land. He made a new friend there, Little Monkey, but he could not fully understand the language of his new friend‟s. The subjects were asked to tell Brother Crab what Little Monkey meant by answering the test questions.
50
Below are examples of the questions shown to the subjects in the Interpretation Task as in Table 3-3 and the Multiple-choice Task as in Table 3-4.
Table 3-3 An Example of the Interpretation Task The subject saw:
The subject heard:
“Laojie” shi shenme yang de jiedao ne?
„What kind of street does „an old street‟ mean? ‟
51
Table 3-4 An Example of the Multiple-choice Task The subject saw: < Scene 1>
The subject heard: < Scene 1>
Pangxie didi zou zai lu shang, yudao xiao houzi. Xiao houzi shou: “Ni di yi ci lai, wo dai ni dao zui renaode laojie guang yi guang, nali you henduo piaoliangde guji.”
Pangxie didi bu zhidao “laojie” shi shenme yisi. Xiao pengyou, qing gankuai gaosu pangxie ba.
„Brother Crab walked along and met the little monkey on the street. The monkey said,
“Let me show you to the most popular old street. There are many beautiful historic spots. Brother Crab didn‟t know what “an old street” means. Please tell him.‟
The subject saw: <Scene 2>
The subject heard: <Scene 2>
“Laojie” shi shenme yang de jiedao ne?
„What kind of street does „an old street‟ mean? ‟ (1) Feichang shouxi „very familiar‟
(2) Henyou lishi „with a long history‟
(3) Henduo lao ren „with many elders‟
52
Both the IT Task and the MC Task presented a designed story scenario and a picture related to the sentence situation. In order to make sure our subjects comprehend the questions and options, designed scenarios shown on the PowerPoint slides were recorded in advance.
3.3 Procedures