3.2.1 Main Texts
There were four main texts (see Appendix B) for both the reading-plus group and reading-only group. Drawing on students’ interest and pedagogical appropriateness, the researcher selected four articles on four different themes from authentic materials on the market. The themes in the four selected articles were: procrastination, gender attraction, shopping, and celebrity, respectively. The purpose for using multiple themes was to control for topic familiarity as an intervening variable in the experiment as well as to control for comprehension floor and/or ceiling effects for a certain topic.
Then, the researcher adapted the four selected articles into four main texts for the study. The four main texts had similar length and similar linguistic difficulty: each main text was around 350-word long, and the ratio of new words in each main text was below 5%, for participants to maintain an enough amount of cognitive space to attend to the message so that they could read with a high level of text coverage and text comprehension (Laufer, 1989, cited in Nation, 2001, p. 145). The readability of
the four main texts was also analyzed using the Automated Readibility Index2, which indicated the U.S. grade level of 8 or 9 needed to understand the main texts. At the end of each main text, multiple-choice comprehension questions were appended to ensure the completion of reading.
3.2.2 Target Words
There were five target words for each main text, twenty target words in total for all the four main texts. The target words embedded in the main texts were not marked in any way, and appeared only once. All of the target words were content words that fit into the contexts of reading materials. Target words were selected based on the criterion that none of them should be familiar to the students prior to the study. To confirm this, several measures for selection of target words were considered. First, Nation’s (1984) vocabulary list was consulted to determine the difficulty level of words in the texts; all the words in the main texts beyond the first 2,000 word list were considered for candidate target words. Second, experienced EFL teachers were consulted to pick up from the candidate words those that have high level of difficulty and high information value to be target words. Third, all the selected target words and the other candidate words as some of the distracters were put onto a checklist; the
2 The Automated Readability Index (ARI) is a readability test designed to gauge the understandability of a text. Its result suggests a U.S. grade level needed to comprehend a text. For example, the result of
“8” indicates the typical reading ability of U.S. eighth-grade students, usually 13-year-old children, to
checklist was used in a vocabulary pretest for the researcher to verify that the selected target words were indeed unfamiliar to the participants prior to the study. Candidate words in the texts that were shown to be unfamiliar in the vocabulary pretest and yet were not selected as target words were replaced with more familiar words or phrases so that learners were allowed to process known information more effortlessly and spend more effort in comprehension of the texts and target words.
3.2.3 Vocabulary-enhancing Exercises
For the reading-plus group, after reading each main text, three vocabulary-enhancing exercises (see Appendix C) were employed on the target words;
in other words, students had three more encounters with each target word. The three vocabulary exercises were intended to encourage conditions of noticing, retrieval, and generation, respectively. The first exercise was to arouse selective attention to particular word forms (Paribakht & Wesche, 1996), which required students to read the list of five target words and locate them by underlining them in the main texts.
This is to draw students’ attention to the target words and thus control for the noticing effect on subsequent vocabulary acquisition and retention. The second exercise was Paribakht and Wesche’s recognition exercise in which students retrieved the meaning of target words elicited from the context in the reading of the main text, and matched them with appropriate definitions. There were more definitions than target words.
This requires association of a written word form with at least one of its meanings, and thus taps students’ receptive knowledge of the target words. As for the third exercise, it was sentence cloze exercise, categorized as Paribakht and Wesche’s production exercise. Students had to complete five sentences using five target words to fill in the blanks. Each sentence was in a new context different from the main text, which encourages generative use of target words. Students needed to take account of the semantic and syntactical features of the target words, and produce an appropriate one to fit into the sentence; this, thus, taps not only students’ receptive knowledge, but also their productive knowledge of the target words.
3.2.4 Thematically-related Texts
For reading-only group, each main text was supplemented by one additional thematically-related text, so there were four supplementary texts (see Appendix D) in total. The supplementary texts were selected from printed or electronic resources thematically related to their respective main texts, compiled and adapted to have appropriate length (around 350-word long) and ratio of new words (under 5%). The same five target words as the corresponding main text were incorporated in each supplementary text. Unlike the one-and-only occurrence of target words in the main text, each target word in the thematically-related passages appeared three times, the same number of times as in the vocabulary–enhancing exercises of the reading-plus
group to avoid the interference of different exposure frequency. At the end of the reading, multiple-choice comprehension questions were again appended so that learners were forced to make an attempt to understand the text meaning and learn the related unfamiliar target words in their process of text comprehension.