The Impacts of Presenting New Words in Semantically-Related Sets on Vocabulary Learning

This paper mainly explores the impacts of presenting new words in semantically-related (SR) sets on vocabulary learning. 38 students from two classes of Grade 2 in Taiyuan Foreign Language School (Senior High School Section) participate in the whole process. The same vocabulary items unknown to all students are taught to them in 4 lessons. The difference between the experimental group and control group is that words are taught in SR sets in the former and in the latter words belonging to the same SR sets are taught in different lessons. Independent Samples T-test is employed to see if there would be a significant difference between the two groups’ performance in the English-Chinese translation tests of those words. The result shows that there doesn’t exit a significant difference between the two groups in SHT test, but in the LT test two weeks later, SU group performs significantly better than SR group. The research findings may provide some enlightenment for foreign language researchers.


Importance of the Problem
Vocabulary learning is important in Second Language Acquisition (SLA).As the famous linguist Wilkins once said, "Without grammar, very little can be conveyed; without vocabulary, nothing can be conveyed" (1972, p. 111).Then there comes a question: As vocabulary learning is of such great importance, which way should teachers adopt to present vocabulary items so as to best benefit the learning?
Presenting vocabulary in SR sets could be seen in many EFL textbooks (Pu 2004, p. 3).Some vocabulary books also put words of a certain category together to facilitate memorizing (e.g.TOEFL Vocabulary Classified).Besides, many writers suggest teaching words in SR sets and have their justifications (Nation, 2000, p. 6).
Although many theorists and practitioners endorse this way of presenting vocabulary, there is not enough empirical evidence to support this position.In recent years, many researchers found that presenting vocabulary in SU sets led to the best outcome of students, while a minority of them found the contrary was better.However, some of these studies used artificial words rather than words from a natural second language.And some provided different words to the experimental and control groups.Besides, few studies in this area have taken senior high school students as subjects; therefore, in the present study, whose purpose is similar to the previous ones, 38 senior high school students were chosen as the subjects and the same new English words were presented to both groups in different manners.

Theoretical Background
Semantic field theory is based on an assumption that the vocabulary of a language consists of many interrelating networks of relations between words, not of a long random list of words.These interrelating networks are semantic fields (Channell, 1981, p. 117).It was psycholinguists who established the concept of semantic field (AlShaikhi, 2011, p. 17) and it was J. Trier, a German Scholar, who brought the semantic field theory to its puberty in the 1930s (Guo, 2010, p. 51).According to Trier's theory, vocabulary in a language is semantically related and forms a complete lexical system, which changes constantly, so we should study the words as an integrated system, only in which a word is meaningful (cited in Wu, 1988, pp. 94-95).In Linguistics: A New Course Book, the relations between words could be divided into different categories: synonymy (e.g.answer-reply, deep-profound), antonymy (e.g.pass-fail, marriedsingle), homonymy (e.g.flour-flower, toe-tow), hyponymy (e.g.vegetable: potato, cabbage and carrot), meronymy (e.g.body: head, neck and leg) and etc. (Liu and Wen, 2006, pp. 120-126).
Interference theory puts forward a different opinion that presenting words in SR sets could not facilitate vocabulary learning process.The interference theory was formulated by McGeoch and McDonald.In their study, it was found that when students were presented with synonyms, they performed poorest (1931, p. 587).According to Robert Waring, this finding, along with others, led to the formation of "Interference Theory", which states that "when words are being learnt at the same time, but are too 'similar' or share too many common elements, then these words will interfere with each Flourishing Creativity & Literacy other thus impairing retention of them" (1997, pp. 261-262).The interference theory provides considerable explanation to the poor performance of the SR group in many studies.

Related Studies
The research on the impact of presenting new words in SR sets on SLA started with Tinkham.He presented related and unrelated artificial words to 20 nonnative adult advanced-level English speakers and found that the participants learnt unrelated words more quickly than related words (1993, pp. 371-380).
Waring in 1997 replicated Tinkham's study in 1993 and the result of his study was the same with Tinkham's (1997, pp. 262-263).
Finkbeiner and Nicol got the same result as the above two.In the study, they also used artificial words as the material for 47 undergraduates, but they presented the same 32 words from 4 categories for both groups.In SR group, related items were blocked into groups of eight, while in SU group, words were scrambled within a block.However, as the 32 items came from 4 categories, it is inevitable that words of the same category may appear in the same block.This situation also happened in Hashemi and Gowdasiaei's study (2003, pp. 369-383).
Instead of using artificial words, Papathanasiou in Greece used 120 English words with 60 related sets and 60 unrelated sets.The result showed that adult beginners performed significantly better in the SU vocabulary tests than in the related ones, while children of intermediate level showed no significant difference between them (2009, pp. 313-322).
The research by Hashemi and Gowdasiaei showed a very different result.60 students ranging from 20 to 30 in two classes in Iran had been taught 100 words and expressions from 13 different lexical sets in different ways in four lessons, respectively.The breadth and depth of their knowledge of the words were tested before and after the instruction.The results showed that students in the SR group achieved better results than students in the SU group (2005, pp. 340-361).
While in China, there were only a few researches in this area and their conclusions all support that of Tinkham.One was conducted by Pu, in which 100 students of the first year in junior high school were chosen as the subjects.50 students were taught 20 SU sets and the others were taught 20 SR sets.Students' knowledge of the words were tested before and after the instruction (2004, pp. 30-41).Another was made by Zhang and Sheng.80 university freshmen were taught 20 new words in SR sets and 20 new words in SU sets.Immediate and delayed tests were given after the instruction (2009, pp. 18-23).
From the review of the related studies, it can be found that in many studies concerning this topic, words for SR group and SU group are different.Some presented the same words to both groups, but in SU group the words of the same SR sets may appear in the same lesson.

Research Question
This study aims to address the following research question: Are there significant differences between presenting new words in SR sets and in SU sets to high school students in the SHT and LT tests?

Sample Description
55 students from two classes of Grade 2 in Taiyuan Foreign Language School (Senior High School Section) in Shanxi, China took the pretest and participated in the material collection part.However, only 38 of those participants attended all the sessions, obeyed all the instructions and remained to the last, with 19 in Class One as SR group (the experimental group) and 19 in Class Two as SU group (the control group).They have learnt English for 7 years and the English proficiency levels of the two groups, as defined by their mid-term English exam scores, were found to be roughly equal.

Material Collection
The paper test has two parts.The first part is about basic personal information---class, sex, years of learning English and student ID number.In the second part, 105 English words are presented with a blank after each one.Students are required to put a "√" in the blank if they know its meaning.This part is designed in the purpose to collect the material for the study.All the words are GRE vocabulary, with at least one third of them appearing on a newspaper for senior high school students, namely Shanghai Students' Post.Besides, these words form 31 SR sets, which belong to five types of relationships---synonymy, hyponymy, homonymy, antonymy and meronymy.The words are presented sporadically on the paper.60 words are expected to be collected as the final material for the study.The material must have two features.First, the material must be words that are unknown to all the students, to ensure that in the SHT and LT tests, if one writes down the meaning of a word, it is not from his/her previous storage of vocabulary, but is learnt from the instruction session.Second, in the material, there shouldn't be any single word which has lost its counterpart(s) in a SR set (see Appendix I.).

SHT and LT Tests
Both tests adopt L2-L1 translation, which is because Nation said it "has the advantages of being quick, simple, and easily understood" (cited in AlShaikhi, 2011, p. 48).One point is given to a student if he/she gives the right Chinese translation to an English word.The right translation is the translation which is taught in class.

Procedures
The study was conducted during normal English class hours and their English teacher was present during all the treatment phase.
Four days before the treatment, students spent 25 minutes doing the papers for the material collection in class.It was found only 39 out of 105 words were unknown to all students.Among the 39 words, some lost their counterparts in the SR set, so some other words were added to make a total of 56 vocabulary items from 22 SR sets.The added words were tested before every lesson to make sure they were new to the subjects.
Four 20-minute lessons were given to each group to teach the 56 vocabulary items with about 14 per lesson.There was a two-or-three-day interval between two lessons.Before each lesson, the vocabulary items to be learnt in that lesson were written on the blackboard, along with their Chinese translation.For the added words, a star mark was written before it and students were asked whether they knew them.Among all the added words, "emolument" was known to one student, so the pair of antonyms containing it---"expenditure" and "emolument" was deleted from the total 22 SR sets, leaving 54 words as the final material.Of the 20 minutes, 15 minutes were used for teaching the new words.The instruction for both classes includes the pronunciation and the meaning of every word, and the sentences containing it.For the SR group, words belonging to the same SR sets were taught together in each lesson (see Appendix II.).For the SU group, words belonging to the same SR sets were taught in different lessons (see Appendix III.).
After the 15 minutes, students were given 2 minutes to review the words.Then the blackboard was cleaned, the SHT test papers were handed out and students were given three minutes to do the L2-L1 translation.After that, students handed in the papers, no matter they finished it or not.Two weeks after the instruction ended, the participants took the LT test with the same 54 items presented sporadically on the paper.

Data Analysis
The data collection includes the data coming from the pretest (their mid-term exam), the SHT test and the LT test.The data for this study are analyzed by SPSS 17.0.
The pretest, which is actually their mid-term exam, was aimed to ensure the rough equality of the proficiency of the two groups.The total mark of the test paper is 150.

Discussion & Conclusion
The present study investigated the impacts of presenting new words in SR sets on vocabulary learning.The results showed that SR group performed better than SU group in the SHT test, but the difference was not significant.On the other hand, two weeks later, the results of the LT test showed SU group performed better than SR group and the difference was significant.
The above result suggests that presenting vocabulary in SU sets may assist the learning of new words better than in SR sets.This result is in agreement with those of previous researches by Tinkham's in 1993, Waring's in 1997and Papathanasiou's in 2009, illustrating that presenting high school students with new words grouped in SR sets impedes rather than facilitates the process of learning.It should be mentioned that the result reinforces the positions of the researchers above since in this study the same new words from natural language were given to both groups in different manners, which previous studies lacked.
The result of the study also adds new testimony to the "interference theory", which believes less vocabulary gains are expected when the words are presented in SR sets.However, the result is not in line with the "semantic field theory", which claims that vocabulary teaching should make associations between SR words.Some guess when presented with new words, intermediate learners only need to add them to an existing storage, so presenting vocabulary in SR sets is best for them (Papathanasiou, 2009, p. 319); however, the finding of this study is against that assumption.
Though psychological researchers provide arguments that words in our brain are related in associative networks (cited in Hashemi and Gowdasiaei, 2005, p. 343), it does not necessarily mean presenting vocabulary items in SR sets facilitates the vocabulary learning process.In this study, presenting vocabulary in this way seems not beneficial to the learners.
Although SU group performed significantly better than SR group in the LT test, SR group performed a little better than SU group in the SHT test.The explanation for this is that in the SHT test with only about 14 words presented each time, there are some cases when words in the same semantic sets share a similar meaning with a nuance of difference, and students are not required to write the difference down, so the Chinese translations are the same for several words, thus elevating students' burden of memorizing.But this situation could only happen in the SHT test.When all the 54 words were presented together in a sporadic manner in the LT test, students in SR group face difficulty recognizing them.

Recommendations for Further Study
There is still much space left for further study.Firstly, there were 55 students from two classes at the beginning of the study, but at last, only 38 remained.The size of the sample is relatively small.If more subjects are involved, more reliable results might be achieved.
Secondly, because of the limited time for the study, the criterion for the choice of words is that the vocabulary items are unknown to all students; however, some words are not often used at an intermediate level.It might have influence on the result of the study.In further studies, the choice of the vocabulary items could be more compatible to the students' English level.
Thirdly, the form of the tests was only about to check students' grasp of the meanings of words, rather than their usages.
Further study could measure the vocabulary depth of students' gains of target words in different ways, such as oral tests, sentence-making and so on.
It is hoped that this study could provide some enlightenment for foreign language researchers.Course designers could organize the words according to their normal use, rather than in contrived SR sets.

Table 1 .
A Comparison of the Scores for the Pretest between SR Group and SU Group

Table 2 .
A Comparison of the Scores for the SHT Test between SR Group and SU Group The results of the LT test show that the mean score of SR group (m=14.2632,std=9.35618) is lower than that of SU group (m=22.5789,std=12.00609)and significant difference (t=-2.381,p=.023<.05) is found between the two groups.
Teachers might try different ways of presenting new words, such as presenting SR items at different times or in different contexts.Students might explore ways to avoid the interference when learning new words.IV blending of synonymy and antonymy (近义反义)混杂 evacuate 疏散 Nearly 3000 nearby residents were evacuated.muster召集(军队)musteran army convene 集合(商讨、开会)convene a summit meeting 纠缠 v. Dog packs constantly badger her about her private life.grouse松鸡，抱怨grouse about excessive taxation 抱怨苛捐杂税 fledgling(刚会飞的)幼鸟，初出茅庐的人 fledgling democracies 新兴的民主国家 III blending of synonymy and antonymy(近义反义)混杂 quandary 困境 George was in a quandary---should he go or not?plight 苦境 the plight of the homeless bottleneck 瓶颈 the bottleneck of production weal 顺境 whether for weal or woe,… The gains offset the loss 得失相当 compensate 补偿 The school apologized and the parents were compensated.One can compensate insufficiency of knowledge with morals; while he cannot compensate moral deficiency with knowledge.respite缓期执行(死刑)respite a murderer adjourn 休会 to stop a meeting or an official process, esp. a trial for a period of time fragile 易碎的、脆弱的 fragile emotion stalwart 强壮的、结实的 a stalwart son grieve 感到悲哀 It is too late to grieve when the chance is past.elate使兴奋 His good grades elate his mom.elated