METHODS
The present study intends to investigate the effect of learners’ SES on learning motivation, use of metacognitive strategies, and L2 vocabulary learning achievement.
Besides, relative explanatory power of SES, learning motivation, and use of metacognitive strategies for L2 vocabulary knowledge will be examined. Thus, the research questions are as follows: (1) Do students of different SES differ significantly in English learning motivation (including motivational intensity and motivation types), use of metacognitive vocabulary learning strategies, and vocabulary learning achievement? (2) What are the explanatory power of learning motivation and metacognitive vocabulary learning strategy for learners’ vocabulary achievement, with SES controlled? The following sections describe the subjects, the instruments, and the data collection as well as analysis procedures.
Participants
The participants included 224 high school students in a vocational high school in Changhua County, Taiwan, where English is learned as a foreign language (EFL context). The particular school was chosen not only because of easy access but its wide diversity of learners’ English proficiency, which might contribute to a better understanding of which factor is more associated with English achievement.
As a vocational high school, this chosen school contains 10 different industrial departments aiming to develop students of different professional skills and knowledge. Nevertheless, students of different departments differ not only in major but in English proficiency, which can be traced back to performance at Basic Competence Test. Generally speaking, those who perform better at BCT are prone to have higher chances of admission to their desired departments in vocational high
47
school. This fact gives rise to a ranking of English achievement among departments within the same grade, despite the same instructional hour per week.
Given this attribute, the present study recruited 224 first-graders from 6 different departments, with three classes considered the most proficient in English and the others least among all the 10 departments. Such a wide dispersion of English ability could be reflected through how much time students devoted to learning English after class. As the participants’ responses to background information questionnaire showed, out of a total of 224 participants, 145 students did not spend any extracurricular hour on English learning. Only one student spent less than an hour on English after class per week; 24 subjects spent one to less than two hours on English per week; 34 participants two to less than four hours; 11 students four to less than six hours; 8 subjects six to less than eight hours; and lastly, only one student devoted more than eight hours to studying English every week. It was expected that such diversity would help find out the connection between L2 vocabulary learning and learner variables such as socio-economic status, motivation, and metacognitive strategy.
Instruments Background Information Questionnaire
The background information questionnaire used in the present study was adapted from Kan’s thesis (2004), where the relationship between motivational intensity and socioeconomic status of learners’ families was investigated. Kan developed the questionnaire based on forerunners’ effort as well. For example, her use of two components – education level and occupation – to rank the status followed Hallinghead’s (1957) two-factor index of social position. In terms of data analysis, education degree was divided into five levels, with the post graduate and above being the top level and allocated five points. College and university level was assigned four points, and high school level three points. Primary school or literate without any
48
schooling was assigned two points, while illiterate level was assigned one point.
Similarly, parents’ occupation was assorted into five categories, from the top five points to the lowest one point. Jobs assigned more points were more professional and skilled than those assigned fewer points. Concrete examples of each occupation category were also given in the questionnaire. (Refer to Appendix A for the details.)
As for scoring, Hallinghead (1957) suggested that SES be calculated through the summation of the higher point of both parents’ education level times four and that of occupation times seven. Based on the summation, SES was further divided into three ranks: high, mid, and low. SES was ranked high if the total fell between 41 and 55, mid between 30 and 40, and low between 11 and 29.
Motivational Questionnaire
The motivational questionnaire consisted of two parts: scales of motivational intensity and motive types. While the former explored learners’ effort, desire, and attitudes toward learning English, the latter aimed to investigate what drove the learners to learn English (see Appendix B). Learners were asked to reply to both scales based on a 4-point Likert scale, ranging from Strongly Disagree (1 point) to Strongly Agree (4 points).
Items 1-11 concerned motivational intensity. They were adapted from Peng’s (2001) and Kan’s (2004) measures of students’ motivational intensity, both of which were translated from Gardner’s (1985) Motivational Intensity Questionnaire (MIQ) of AMTB and modified to suit Taiwanese contexts of varied learning levels. In addition, in order to facilitate data analysis and prevent confusion in respondents, all the negatively worded questions were transformed into positive ones in the current study.
For example, both Peng and Kan inquired learners on how they perceive their English assignment with the statement, “When I do my English homework, I often feel distracted and just skim over it.” In this study, such a statement was changed into
49
“When I do my homework, I concentrate in the process and do my best.”
The second part of the motivational questionnaire comprised items concerning motive types. There were 34 items in total, namely, items 12-45, categorized into four subsections: intrinsic motivation, extrinsic motivation, amotivation, and requirement motivation. The first three sections were based on the construct, Self-Determination Theory (Deci & Ryan, 1985), whereas the last section was founded in Warden and Lin’s (2000) suggestion that requirement motivation accounts for the language learning motivation of Taiwanese students to a great extent. The instruments measuring these attributes were adapted from Language Learning Orientations Scale
－Intrinsic Motivation, Extrinsic Motivation, and Amotivation Subscales (LLOS-IEA) by Noel et al. (2000) and a mini-questionnaire on requirement motivation generated by Peng (2001). The purposes of this study are similar to Kan’s (2004); therefore, Kan’s translation of LLOS-IEA and her modification of Peng’s requirement motivation questionnaire were adopted. Table 2 displays the organization of the motivational questionnaire and distribution of these items. The complete motivational questionnaire is shown in Appendix B.
Table 2
Organization of the Motivational Questionnaire and Item Distribution Part 1: Motivational Intensity Part 2: Motive Types
Content
50
Metacognitive Strategy Questionnaire
The third questionnaire was developed by the researcher with reference to Gu and Johnson’s (1996) vocabulary learning strategy questionnaire and Schmitt’s (1997) vocabulary learning taxonomy, both of which contain items tapping learners’
metacognition in vocabulary learning. As mentioned in chapter two, the strategies proposed by the two studies coincidentally represent two different but complementary dimensions of vocabulary learning. Specifically, while Gu and Johnson’s strategy categories of Selective Attention and Self-Initiation pertain to the encounter and discovery of new vocabulary, Schmitt’s Metacognitive Strategies concern consolidation of the learned vocabulary knowledge. Therefore, the present study combines the metacognitive strategies suggested in both studies, and renames Schmitt’s category as Consolidation in order to explicitly demonstrate its function in vocabulary learning as Gu and Johnson’s does.
In detail, strategies in Selective Attention are employed to make judgments about whether the newly encountered word is worth attention and further learning. In other words, Selective Attention concerns with judging whether there is a need to guess, look up, or memorize the new vocabulary. Items 1-7 reflect learners’ use of these strategies.
Also applied to the learning of new vocabulary are Self-Initiation strategies. This category highlights learners’ active exploration of English words after school other than accomplishing required assignment. The response to this category reveals whether learners take active and autonomous control over their building of English lexicon, that is, whether they can make use of extracurricular resources to expand their word bank. Items 8-12 aim to find out the frequency of self-initiated vocabulary learning. Items on both Selective Attention and Self-Initiation were translated by the researcher herself. And some of Gu and Johnson’s items were adapted to meet the
51
participants’ learning situation and to reach higher clarity. For example, in item 8, the scope of self-initiated learning was expanded from “other readings” to “all kinds of resources” given the popularity of multimedia learning resources nowadays. Also, in item 10, the researcher replaced the obscurity of “things” with “English vocabulary”
to avoid the participants’ distraction from lexical learning. In addition to translation and rephrasing, all of the negatively-worded items were changed into positively-worded statements, for the same reason in adapting motivational questionnaire. For instance, in item 9, the researcher changed the original wording “I wouldn’t learn what my English teacher doesn’t tell us to learn” to “Besides required assignment by teachers, I would also actively learn English words that are not in the curriculum”; item 10, from “I only focus on things that are directly related to examinations” to “I don’t just learn English words that are related to examinations”;
item 11, from “I wouldn’t care much about vocabulary items that my teacher doesn’t explain in class” to “I don’t just learn English words that my teacher explain in class.”
Lastly, the category of Consolidation probes into how learners plan and monitor their consolidation of words that have been introduced. Items on this category were adapted from Hu’s (2007) translation of Schmitt’s metacognitive strategies, yet with the deletion of one of Schmitt’s items that is not considered to adhere to the function of consolidation of learned word knowledge: Skip or pass new word. Also, two additive items were included in this category with reference to Hu’s – items 16 and 17 – so as to make the investigation more comprehensive. Table 3 shows the item distribution in the metacognitive strategy questionnaire.
The participants were asked to rate each statement on a 4-point Likert scale, from Never Use This Strategy (1 point) to Always Use This Strategy (4 points). Therefore, given 17 items in total, students’ scores may range from 17 to 68. Appendix C shows the questionnaire investigating learners’ use of metacognitive vocabulary strategies.
52
Table 3
Item Distribution in the Metacognitive Strategy Questionnaire
Category
Number of items
Item number The lowest score The highest score
Selective Attention 7 1-7
17 68
Self-Initiation 5 8-12
Consolidation 5 13-17
Vocabulary Knowledge Test
The vocabulary tests used to assess participants’ vocabulary knowledge are adapted from Nation’s Vocabulary Size Test from his webpage (Nation, 2007).
Nation’s VST originally contained five levels — the 2nd 1000 word level, the 3rd 1000 word level, the 5th 1000 word level, the academic word level, and the 10th 1000 word level — which aimed to determine English learners’ vocabulary size. Recently, the test for the 1st 1000 word level also became available.
The present study includes the first two levels test — the 1st 1000 word level and the 2nd 1000 word level — under the consideration of the participants’ proficiency level. Since junior high school graduates should possess a lexicon of the most frequent 1,000 words, the 1st 1000 word level test should be appropriate to begin the assessment. On the other hand, given the fact that vocational high school graduates are expected to reach the 4,000-word level, it is also considered sufficient to have first graders to take the test of the 2nd 1000 word level.
According to Nation’s design, each test contains 60 representative words of that frequency level, organized into 10 clusters based on parts of speech, six words in a cluster. Next to each cluster are three definitions of three random words out of these six words. Test takers need to choose the corresponding words out of the six options to match the three given definitions. Thus, each test consists of 30 questions in total.
53
Each correct answer is given one point, making 30 points at most. Learners are considered to master a particular level – know enough words of that level and may move on to learning lexis in the next level – as long as they answer at least 27 out of 30 questions correctly, which proclaims that they have acquired 90% of the words in that frequency list.
However, the present study does not intend to find out which level of vocabulary the participants have mastered. Rather, the 1st and 2nd 1,000 word tests are combined together and used as an achievement test without level boundary in the present study.
In other words, the participants’ performance on this combined vocabulary test does not entail their degree of mastery of vocabulary at certain frequency level, as Nation postulated. Given 60 questions in total and one point for each correct answer, the combined vocabulary test yielded scores from 0 to 60 points.
More noteworthy is the language version of the two level tests in this study. They were all bilingual, with three definitions in each cluster written in Chinese rather than English. Both tests were directly adopted from Nation’s website (http://www.victoria.ac.nz/lals/staff/paul-nation.aspx). Although Nation contended that in the monolingual version the English used in definition are at higher frequency level than the words to be tested and is not supposed to hinder test takers’ answering process, the researcher decided to employ bilingual tests, in order to make the tests more subject-friendly and to avoid any possible confounding effects caused by English, a foreign language to the participants. Appendix D shows the combined bilingual version of 1st and 2nd 1,000 word level tests in use.
Procedure Data Collection Procedure
Before the formal study, a pilot study was conducted. A group of 74 students in the same vocational high school as that in the formal study were chosen based on
54
convenience sampling. They were asked to complete all the three questionnaires and the vocabulary test as fast as possible so that the researcher could reach an understanding of how much time students would need to complete the questionnaires and the test in the formal study. Besides, their responses to the motivation and strategy questionnaires were submitted to statistical analysis of reliability. Students were also asked to give feedback on the clarity of item wording.
To measure the internal consistency of the following sub-questionnaires － motivational intensity, intrinsic motivation, extrinsic motivation, amotivation, requirement motivation, selective attention strategy, self-initiation strategy, and consolidation strategy － Cronbach’s alpha for each of these 8 subscales was calculated. Following conventional standards, those whose alpha was over .70 were considered reliable and would not be modified. If the alpha level was below .70, further modification would be made based on item analysis.
The formal study took place in December in 2011 after the school midterm examination. The researcher asked five of her colleagues to help invite their classes to join the research. The directions for implementation of the study was fully explained and the announcement to be made to student participants before distribution of the instruments was also demonstrated for the instructors. The participants were told that there was no right or wrong answers for the questionnaires and that their responses to the items would not affect any of their course scores. They were encouraged to answer forthrightly. Moreover, to lower participants’ affective filter and elicit subsequent honest and unreserved responses, the time limit for completing the instruments was set based on the feedback of the participants in the pilot study.
Data Analysis Procedure
The data were keyed in computer and then analyzed via SPSS. In addition to descriptive statistics, MANOVA was run to measure if students from different SES
55
groups differ significantly in both motivational intensity and motive types. One-way ANOVA was performed to reveal if students from different SES groups differ significantly in use of metacognitive strategies and vocabulary achievement.
More importantly, SES, motivation (both intensity and types), metacognitive strategy, and vocabulary achievement were submitted to hierarchical regression, with SES controlled in the first phase and motivation plus metacognitive strategy serving as the second-phase independent variables, so as to illuminate whether motivation and metacognitive strategy significantly predict vocabulary achievement in addition to learners’ socio-economic status.
56