In the field of second language acquisition, researchers often employ either qualitative methods or quantitative methods to assess L2 learners’ competence and performance of the target language (Larsen-Freeman and Long 1991:11). Researchers who adopt qualitative methods do not seek to test a hypothesis; rather, they observe L2 learners’ linguistic development in naturalistic settings. Besides, the number of subjects is small and the study is carried out longitudinally. A quantitative study, on the other hand, aims at testing a hypothesis by recruiting a larger number of subjects from different groups to perform certain linguistic tasks in experiments. The data are obtained only from the experimental sessions rather than in an extended period of time.
Compared with a qualitative approach, a quantitative approach has the advantage of collecting data from a greater number of subjects within a shorter period of time.
Moreover, the quantitative data allow researchers to make a cross-sectional comparison and generalize developmental patterns (Bardovi-Harlig 2000:202-203).
For a quantitative approach, two major types of data can be collected. Production data,
level (H) groups in the present study.
gathered from structured exercises, elicited translation, interviews, and so forth, can reveal L2 learners’ linguistic performance during the process of second language acquisition. Comprehension data, elicited though tasks such as error recognition and grammaticality judgments, are able to uncover L2 learner’s competence (Larsen-Freeman and Long 1991:27-35).
To investigate L2 acquisition of the English progressives, some previous research adopted comprehension tasks to examine L2 learners’ competence (cf. Chin 2006, Gabriele and Canales 2011), and others chose production tasks to explore L2 learners’ performance (cf. Huang 1999, Muñoz and Gilabert 2011, Chuang 2012).
However, few of them employed both types of tasks to collect data. Given this, the present study incorporated both comprehension and production tasks in order to gain a better understanding of the acquisition of the English progressives by Chinese EFL learners.
There were two tasks in the present study: an acceptability judgment task and a blank-filling task. The two tasks were designed in written form to test learner’s comprehension and production, respectively. Details of these tasks are provided in the following subsections.
3.2.1 The Comprehension Task
To test our subjects’ comprehension of the English progressives, an acceptability judgment task (AJ Task) was developed. The subjects were instructed to rate the acceptability of sentences on a four-point scale2, with 4 being very acceptable, 3 acceptable, 2 somewhat acceptable, and 1 unacceptable. Two examples of the task are shown in Table 3-1. (Please refer to Appendix A for the directions and complete test
2 We chose even number instead of odd number of points in order to make the participants give careful consideration to each question.
items for the task.)
Table 3-1 Two Examples of the Acceptability Judgment Task
A. He is complaining about how messy his roommate is. 4 3 2 1 B. The little boy is jumping up and down on the sofa. 4 3 2 1
Sentence A dealt with the combination of the progressive marker and an agentive activity verb. Based on the results of the previous studies and the existence of such combinational pattern in the learners’ L1, a high acceptability rate for this category was expected. Sentence B tested the progressive marker with an agentive semelfactive verb. Though this category did not receive much attention in the literature of second language acquisition, a higher score was also expected because of positive L1 transfer.
As mentioned in Chapter Two, the present study proposed a new classification of verbs which consisted of ten subtypes. In the AJ Task, there were two trial sentences for each subtype. The trial sentences were taken from previous literature, dictionaries and English magazines. The task contained 25 trial sentences (10 verb subtypes × 2 sentences＋5 fillers＝25) and they were randomized. Table 3-2 presents the design of the AJ Task.
Table 3-2 Test Design of the Acceptability Judgment Task
Note. For simplicity, we followed the convention (e.g. Li 2006, Housen 2000, Chuang 2012, Gabriele and Canales 2012) and adopted the three-letter abbreviation for each verb type.
Since Chinese exhibits the combination of the progressive marker with activities, semelfactives, and accomplishments, it was predicted that the subjects would perform better on the three types. Moreover, a better performance on agentive verbs of all categories was expected.
3.2.2 The Production Task
Besides examining the subjects’ comprehension of the English progressives, we also designed a blank-filling task (BF Task) to test the subjects’ production. In the literature, oral interviews and picture-elicited narrative tasks have been conducted to investigate L2 learner’s production of progressives (cf. Huang 1999, Muñoz and Gilabert 2011). However, these types of tasks often generate an imbalanced number of tokens from different verb types (Bardovi-Harlig and Reynolds 1995, Bardovi-Harlig
2000:201). For example, the participants use the progressive aspect with activity verbs much more frequently than with other verb types. Results like this are sometimes hard to account for. Learners may not have acquired progressives with other verb types or they have acquired but simply don’t use it. To solve these problems, the present study adopted a controlled task which included two trials for each verb subtype.
In the BF Task, a picture along with guided words was provided in each trial.
The subjects were asked to describe the current condition of the person or the thing in the pictures by filling in the blanks. Table 3-3 presents two examples of the BF Task.
(For the complete version of the task, please refer to Appendix B.)
Table 3-3 Two Examples of the Blank-Filling Task
The boy (build) a sandcastle.
The water in the teapot (boil).
Picture A was intended to elicit the agentive accomplishment and Picture B was for the non-agentive activity. The subjects had to choose a suitable verb form to depict the pictures. There were 25 trials for the BF Task (10 verb subtypes × 2 sentences＋5 fillers＝25), as shown in Table 3-4.
A. B.
Table 3-4 Test Design of the Blank-Filling Task
It was expected that the subjects would use the progressive aspect more to depict activity, semelfactive, and accomplishment situations. Also, agentive verbs of all the types would receive the progressive markings more than non-agentive verbs.