This section is devoted to the procedures of the pilot study and the formal study as well as the scoring and statistical analysis adopted in the present study.
3.3.1 Pilot Study
The pilot study recruited subjects who have passed different levels of General English Proficiency Test (GEPT). Those who passed elementary level belonged to the low group (L), and those who passed intermediate and high-intermediate fell into the mid group (M) and the high group (H), respectively. There were two subjects of each level in the pilot study.
A grammaticality judgment task (GJ Task) and a blank-filling task (BF Task) were conducted. For the GJ Task, the subjects were asked to judge whether the sentences were grammatical or not. One concern with the GJ Task was that when the participants rejected a sentence, it was difficult to tell whether the learners were rejecting the target structure or some other parts of the sentence. Therefore, the participants were asked to correct the sentences which they considered ungrammatical.
Error correction could help the researcher to arrive at a more precise analysis of the learners’ answers. For the BF Task, pictures featuring people or things under certain conditions along with guided subjects and verbs were given. The participants had to describe each picture by filling in the blank based on the information provided.
In the comprehension task, each correct answer to the question received one point while a wrong answer received zero. Under the circumstance that a participant rejected a sentence but corrected the parts other than the progressive morpheme, one point was still given. In the production task, each blank that was filled with the progressive aspect received one point and others got zero.
The major findings of the pilot study are summarized as follows3. First, there seemed to be positive and negative L1 transfer. Chinese exhibits the combination of progressive marker zai with activities, semelfactives, and accomplishments, and the participants performed better on the items that involved the English progressive marker with the three types of verbs. On the contrary, states and achievements cannot co-occur with the progressive marker in Chinese and the participants did worse on the items of these two verb types. Second, the acquisition of English progressive aspect by EFL learners seemed to begin with activities and accomplishments. Third, the participants performed better on agentive verbs than on non-agentive verbs in both tasks. Fourth, proficiency was found a factor affecting our learners’ performances.
3 The results of the pilot study are presented in Appendix C.
Learners of the M and H groups gave more correct responses than those of the L group. Finally, the task effect was found evident. All the groups performed better on the comprehension task than on the production task and the L group showed the greatest differences between the two tasks.
There were some limitations of the pilot study. First, since the subject pool was rather small, it was not representative enough to make a generalization of the acquisition of the English progressives. A larger pool of participants was needed to substantiate the results. Second, error correction of the GJ Task revealed that the participants preferred simple aspect over progressive aspect in some cases. It was not clear whether they rejected the sentences because of the incompatibility of English progressive with certain types of verbs, or the combination was just less satisfactory in sentences without context. Therefore, the GJ Task needed to be revised to further explore the subjects’ knowledge of English progressives.
3.3.2 Formal Study
Given the inadequacies of the pilot study, the formal study replaced the GJ Task with the acceptability judgment task (AJ Task) to test the learners’ comprehension of English progressives and recruited more people to participate in the study.
The subject pool consisted of 80 non-English majoring freshmen in National Taiwan Normal University. There were 20 people in the L group, 20 in the LI group, 20 in the HI group, and 20 in the H group. The researcher first obtained the permission from the teachers of the Freshmen English classes to give a brief introduction of the study in the classes and then recruited students who were interested in participating in the study. Before taking the two tasks, each of the
participants was asked to sign a consent form4. The participants were assured that their answers to the tasks would be used only for academic research and the personal information gathered during the study would be kept confidential.
The study was carried out in the subjects’ classrooms. One copy of the questionnaires was distributed to each subject. To ensure the subjects’ understanding of how to answer the questions in the two tasks, examples of the tasks were provided on the copy. Also, clear and succinct Chinese instructions were given before the trial session started. The questionnaire contained a blank-filling task and an acceptability judgment task. The subjects were encouraged to answer all the questions without leaving any blanks. To circumvent potential practice effects, the subjects were asked to finish the BF Task first and then preceded with the AJ Task. It took each participant approximately 30 minutes to complete the two tasks.
3.3.3 Scoring and Statistical Analysis
After the data were collected, all the responses were scored and processed by SPSS software for statistical analysis. There were target sentences and fillers in both tasks but only responses to the target sentences were scored and analyzed. For the AJ Task, the scores were given based on the subjects’ ratings. Therefore, the score for each question ranged from one to four, as illustrated in Table 3-5.
Table 3-5 Scoring of the AJ Task
Acceptability Rating Point
Very acceptable 4 4
Acceptable 3 3
Somewhat acceptable 2 2
Unacceptable 1 1
4 The consent form is provided in Appendix D.
For the BF Task, the answers of the verb forms were analyzed. Since the subjects were asked to describe the current condition of the person or thing in each picture, the answers were expected to be in the form of present progressive. For instance, in the question “The water in the teapot _______(boil),” the correct verb form should be “is boiling.” Table 3-6 illustrates the scoring of the BF Task.
Table 3-6 Scoring of the BF Task
Type Example Point
Present progressive
The water in the teapot
is boiling. 1
Spelling mistake is boilling. 1
Past progressive was boiling. 0.5
S-V disagreement are boiling. 0.5
be-omission boiling 0.5
Non-progressive boiled. 0
boils. 0
As can be seen from the above table, there are three kinds of scores for the answers. First, answers expressed in present progressive received one point. Answers expressed in present progressive but with spelling mistakes on the verb forms received one point as well. Second, answers expressed in past tense, with a wrong be verb, or without a be verb received 0.5 point. Though they were grammatically incorrect, these answers showed that subjects were aware of the concept of progressive, which is to describe something in progress. The misuse of tense or the be verb might simply result from negative L1 transfer since Chinese does not have a tense system and does not require subject verb agreement like English and exhibits the omission of the copula verb. Third, answers that were not expressed in the progressive aspect received zero point.
All of the scores were processed by the SPSS software for statistical analysis.
The descriptive statistics was run to calculate the means and standard deviation of the data. One-way analysis of variance (ANOVA) and post hoc Tukey were executed to examine the extent of differences among the four proficiency groups as well as the verb types. Furthermore, t-test was performed to examine the transfer effects and the agency effects.