A Pilot Study On Classroom Observation Education Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Chapter 4

This chapter describes a pilot study conducted after the preliminary study and before the main study. It begins by a brief introduction given in section 4.1, followed by discussions on the sample selection, observation schedule and pre and post session interviews in section 4.2. Section 4.3 covers the classroom observations. Gesture coding, including the transcription, coding problems and policy, and results of the coding are included in section 4.4. Discussion and conclusion focusing on classroom observation and metaphoric gestures are covered in section 4.5. The chapter ends with the implications for the main study explored in section 4.6.

4.1 About the Pilot Study

This pilot study built on a previous small scale preliminary study (chapter 3). The aims were: (a) to experience real time classroom observation; (b) to test out and improve the observation schedule and interview questions; (c) to explore whether Taiwanese music teachers used metaphoric gestures and if so, what these metaphors were and where they were used; and (d) to test out and develop metaphoric gesture identification procedures.

The focus was limited to music sessions in junior high schools because in the preliminary study, more metaphors were found from the junior level than from the elementary level. The data of this pilot study were transcribed from three general music sessions taken by Wang, a music teacher in a junior high school in Taiwan.

4.2 Before the Observation

4.2.1 Sample Selection

Situated in a middle to upper-middle class suburb of a major northern city, Wang's school was founded in 1988, with 4,268 students and 113 classes in 2006. It is considered to be a big school, relative to the official average of 1,299 students per junior high school in Taiwan for the school year 2005-2006 (Department of Statistics, Ministry of Education of Taiwan). The data were collected in general music classes with students aged between 12 and 14. Each class contained around 35 students, and lecture was the only type of interaction observed. The results of the preliminary study showed that metaphor density may differ because of teaching content. Therefore the sessions observed for this pilot were limited to music theory, music history, and music appreciation.

Wang (a pseudonym) had received her MA in musicology four years previously and since then had been teaching music in the same junior high school. Being very open-minded about taking part in research and being observed, Wang was one of the very first music teachers I contacted for the preliminary study. Since then she had been helpful in answering my questions and providing me background information about the situation of music education in junior high school level in Taiwan. After the preliminary study, I emailed her to ask her permission for me to enter her classroom and video-record a couple of sessions, and she agreed to participate.

Music lessons form part of the Arts and Humanities course category in junior high schools in Taiwan (see section 2.4.1), and therefore music, arts, and performing arts share one textbook. "The arts section talks about using different colours to represent the four seasons, and the music section talks about Vivaldi. So it occurred to me that it's a good opportunity to give a lecture on the Baroque era and introduce Vivaldi's The Four Seasons to them," (Wang, pilot interview 2). At the time when this report was written, MOE of Taiwan did not provide one "standard" version of the textbook, and schools were free to choose the version they preferred.

Wang therefore regularly designed the content of her lessons. "The textbook mentions nothing about the Baroque era so I make my own PowerPoint slides to help students establish the background knowledge" (Wang, the same interview). Teachers are allowed to re-arrange the order of the teaching contents and put related things together.

In some schools, music teachers have to teach performing arts, but this was not the case in Wang's school, where there were three individual teachers for the three sub courses: music, arts, and performing arts.

4.2.2 Observation Schedule and Interviews

After Wang agreed to participate in the study, I explained to her about the research through email. She understood that the research was about classroom discourse and therefore suggested me to set the observation time at a week after the mid-term exam, when she was about to introduce the Baroque era to her students. One month before the classroom observation, Wang and I met to discuss which classes to observe, including when I should arrive, where I should sit and how best to set up the recording facility. In addition, Wang roughly explained what kind of classroom activities would be involved and what she intended to teach. Wang was told that the study was about classroom talk and the idea was to observe a session with as much talk involved as possible. She therefore suggested sessions mainly covering music history and music appreciation. This meeting is classed as pilot interview 1 (for more details, see Appendix B).

A real-time observation schedule was designed, with a purpose of linking the classroom activities and metaphor use. As can be seen from the schedule (see Appendix C), both start and end times of the activities needed to be specified and during each activity, tallies of metaphors and gestures needed to be made, so that I could get a basic sense of where clusters of metaphors and gestures most often occurred (though in the event it proved almost impossible for me to count them during classes, something which will be discussed later in 4.6.1). In addition, the observation schedule covered: organisation of the class, materials, and musical instruments used, and the language used by the teacher and students. A general impression of classroom atmosphere would be noted as H (high), F (fair), or L (low) to see if classroom atmosphere is related to metaphor use.

A follow-up interview (pilot interview 2) with Wang was conducted after observing the three sessions. The face-to-face semistructured interview took place in Wang's music classroom during her break, lasting 42 minutes. Again, I was permitted to record it, and at the same time made notes while Wang was talking. A list of information or questions I planned to seek or ask, translated from Chinese (the language used during the interview) into English is appended (Appendix D). In the interviews, information was sought about Wang's educational background and working experience, and questions covered how Wang prepared for the class and aimed to explain new concepts, how Wang thought about metaphors and gestures, and if she used them to help her teach. Mandarin Chinese was used throughout the interview. Notice that the numbers are for the convenience of writing up the report; the questions were asked without any specific order.

During the interview with Wang, the above topics were covered. To Wang, metaphor helped the students to connect music and their daily life experiences. She gave an example of the birds, dogs, spring wind, and summer thunder in Vivaldi's The Four Seasons. However, she did not usually interpret music in her own way when introducing or describing it-"I only interpreted the music in the way which the composer marked on the music manuscript" (Wang, pilot interview 2).

Although overall the interviews went smoothly, it was realised that some questions were too general and more questions about metaphor and gestures were needed. Details are discussed in 4.6.1.

4.3 During the Observation

For the first session, I entered the music classroom with Wang around 15 minutes before the session began. Wang told me that because of the limited numbers of music classrooms, music teachers in her school had to take turns using them. It was Wang's turn to have one of the music classrooms that semester and therefore she was allowed to stay in the room waiting for the students to arrive. The extra time proved useful from the point of view of setting up the video equipment.

The main equipment in the music classroom included a piano, an electronic piano, a blackboard with blank musical staff, an LCD projector, a DVD player, and a few classical musicians' portraits and illustrations of musical instruments on the walls, etc. Wang had to provide her own laptop.

Figure 4.1 illustrates the layout of the classroom where the three observed sessions took place. The two black circles on the top photo indicate where the video recorder was placed and where I sat. The video recorder started to record when the students started to walk into the classroom. It was not turned off until the class was dismissed and the classroom was empty.

Neither the teacher's nor the students' seats were moved. As a nonparticipant researcher, I tried to keep the classroom the way it was without me. I sat at the back of the classroom next to the video recorder, to take field notes which might be helpful during the preparation of the transcripts. Photos were also taken before or/and after sessions. The video recorder started to record when the students started to walk into the classroom. It was not turned off until the class was dismissed and the classroom was empty.

Figure 4.1. Pilot study: Music classroom setting.

For most of the time the students could not see me during the classes. However, there was one time, while Wang corrected a student's fingering while playing the recorder, another student kept turning and gesturing "YA" (the V sign, with his index and middle fingers of his right hand raised and the remaining fingers clenched, palm facing outwards; a very popular gesture in Taiwan when being photographed) toward the video recorder. Wang noticed it. She asked the student to stand up and introduce himself to the video recorder, and encouraged him to give a solo performance for two bars, and then the session continued. This was the only interruption caused by the researcher's presence during the sessions I observed.

The lecture structure of the three sessions I observed was quite similar. It was because Wang designed and followed her own teaching syllabus for each year of the students, and also because the three sessions I observed were all in the same year (year seven). The session started with playing the recorder. Wang reviewed the piece she had taught in the previous session with the whole class and then selected a few students to stand up and play individually, in order to discover how much students had learned (Wang, pilot interview 2). Then she taught one new piece by demonstrating and playing with the whole class. After the recorder playing, the lecture part "episode" (Lemke, 1990) started. Wang began by introducing the recorder ensembles: bass, tenor, alto, soprano and sopranino recorders. Then she introduced some important eras in musical history, with a focus on the Baroque. To this end, Wang gave students some background knowledge about the characteristics of the music, musical instruments, and some famous musicians, before introducing Vivaldi and his concerto, The Four Seasons.

Generally speaking, the observation schedule proved easy to mark. Keeping records based on classroom activities made it easy to recall what happened in the sessions, and there was enough time to keep records for most of the categories, except for the numbers of linguistic metaphor and gesture used. Secondly it became clear that some categories needed to be further specified-this will be discussed later in 4.6.1.

4.4 Gesture Coding

McNeill's scheme (see sections 6.3.2 for a review of the literature and 6.3.3 for a discussion, on gesture classification systems) applied in this study required asking what meanings and functions a gesture possessed. In other words, the categories were not based on just one facet of a gesture. For example, iconics and metaphorics were more semantically oriented, while deictics were more pragmatically oriented. Because of this, each category was not treated as discrete or mutually exclusive, but as having features that may be present in varying degrees, and possibly in combination. Thus, the ultimate goal of gesture coding is to identify the extent to which each feature is present, rather than classify the gestures (Eisenstein & Davis, 2004; McNeill, 1992). In this pilot then it is very important to set up a coding policy to indicate when to categorise a gesture by its meaning and when by function. More details are discussed in 4.4.3.

Two coders were involved in gesture coding in order to test out McNeill's procedure and increase the reliability of the study. The other coder, a graduate student in the Department of Educational Studies at the University of York, was a native speaker of Chinese with some knowledge of metaphor having conducted her MA research on the topic. McNeill's three main gesture categories-deictics, iconics, and metaphorics-were introduced before I asked her to categorise the gestures which Wang used. Due to time limitation, only a part of the listening to The Four Seasons from Wang's first session was selected. The part was chosen for its dense distribution of gestures. It lasted for five minutes and ten seconds, with 35 gestures identified by me previously. More details about gesture coding are discussed in 4.4.4.

4.4.1 Data Selection

Due to the limited time available, only one session was chosen for gesture coding. The reason for choosing the first session was that although the structure of the three sessions was similar, the first session was more complete and covered all the activities from recorder playing, a lecture on the Baroque era, to music listening, with a more even allocation of time, than the other two sessions. Types of activities matter here because if numbers of gestures differ in different types of activities (and it seems so from the results), involving more activities in the data can reduce the risk of any possible gesture loss.

4.4.2 Transcribing the Data

Both gestures and speech were transcribed. Speech was transcribed fully from the videotape in Chinese. The transcription of gestures included three steps: (a) identify the movements that were gestures (here gesticulations); (b) identify the stroke of the each gesture; and (c) locate the boundaries of the gesture phases in the relevant part of the phonological transcription.

4.4.3 Coding Problems and Policy

The results of the gesture categorisation from the two coders were compared. Unlike deictics, which were all agreed by both coders, iconics and metaphorics seemed to constitute the more problematic categories. One of the examples was where Wang lifted her left hand at shoulder height with her palm facing up and wiggled her middle, ring, and little fingers fast in turn when saying the word zhuangshiyin ("trill"; literal translation: "decoration notes"). Although both coders agreed that Wang was holding an invisible violin and playing the trills, we nevertheless coded it differently. The other coder coded it as metaphoric, because the referent of the gesture, zhuangshiyin, was interpreted as a linguistic metaphor itself. I coded it as iconic because the gesture in fact was representing what was explicit in the accompanying speech and therefore had a close relation to the gesture. This helped develop one of the coding policies later-gestures accompanying metaphors in speech were not definitely considered to be metaphorics.

Another problem related to the coders' differential focuses and interpretations of the same gesture. Wang said, "The thunder came fast and went fast. Then it became quiet again." In the first sentence she began by moving her left arm from left to right with the palm facing up and fingers curling and then moved the arm back to the left without changing the shape of the palm. For the second sentence, she turned over her left palm to face the ground with straight fingers and pressed down the palm a little bit in the air. The different coding concerned the first gestural phrase. When I coded them, I focused on the movements of the arm, which was moved from left to right, and then back to the left, which seemed to indicate "came" and "went" in speech. Therefore the gestures were coded as iconic (rather than metaphorics, which was clarified later). On the other hand, the other coder focused on the shape of the palm, which changed from a palm with curled to straight fingers. She interpreted the curled fingers as representing the "fast" in speech, while the straight fingers represented the "quiet," and so she coded them as metaphoric. What was stunning was that none of us thought about interpreting the same gesture each other's way before having the discussion. Our agreed position was to class the gestures as iconics, because both of us agreed that it was more common to see such gestures appearing along with "came" and "went" than "fast" and "quiet."

In addition, what the other coder received before coding might have affected her. She used my transcription of gestures while categorising the gestures. Although the gestures were described in a descriptive language, it was found later that some of the transcription was in fact subjective. For example, descriptions such as "the right hand drew a shape of flash light" or "the left hand indicated calmness" were already interpretations and they could be misleading to the other coder.

These problems not only predicted what may happen during the coding process for the main study, but also helped develop the coding policy to be used, which is discussed below.

Identifying Gestures

In this study, "gesture" specifically refers to gesticulation. Any fingering the teacher used to demonstrate how to play the recorder, or the conducting gestures often used while the class was playing the recorder were beyond the scope of this study and excluded. The interpretation of Wang's gestures was made from the researcher's perspective, and it is worth noting that this might differ from the interpretation from the speaker's or the addressee's perspectives (A. Cienki, personal communication, June 3, 2008).


Metaphoric gestures were defined as gesticulations which present a more abstract referent in terms of a more concrete image and engage a cognitive process of understanding one thing in terms of something else. This definition dovetails reasonably well with Lakoff and Johnson's conceptual metaphor theory, the theoretical framework on which this present study was built, and at the same time does not contradict the Pragglejaz definition of metaphorically used lexical items applied in the study (discussed in section 3.3.2 and later in 6.4.2).

One example is when Wang said gangqin de yinse yue lai yue xizhi ("the timbre of the piano becomes more and more delicate," and gestured using a round and half-open palm facing up accompanying the word "delicate." Here, the gesture carries the dual structure required by a metaphoric, in which the representation of the delicacy of the timbre (a more abstract referent) by the gesture is presented as what appears to be an image of a bud waiting to open (a more concrete base).

Deictics versus Metaphorics

Both McNeill's deictics and iconics were re-defined for the present study. According to McNeill, abstract pointing gestures which imply a metaphorical picture are also included in the category of deictics. For example, gestures were used in my data to point at an existing physical place, but they referred to as an abstract concept of where the speaker had been before. When Wang said "Did we just say that decoration is popular in Baroque era," she raised her left index finger to point to the "idea" which she just mentioned in the same session. Pragmatically speaking, these gestures were pointing gestures (deictics), but semantically speaking, the place which the gestures pointed to was interpreted as somewhere else based on the speech context. In other words, such gestures engaged a cognitive process of understanding something (the physical place which the gesture pointed at) in terms of something else (the actual space or idea which the interlocutor talked about previously) and therefore were classified as metaphoric.

Iconics versus Metaphorics

Gestures in a context where the Vehicle of a metaphor is explicitly flagged both by the hands and by language were classified as iconic. That is, if a teacher says "music is a container" and gestures a container, the gesture will be categorised as iconic rather than metaphoric because the gesture may accompany metaphor, but itself represents the literal form of the word "container"; however, if the same gesture accompanies the sentence "we can feel the sadness in his music", it will be classified as metaphoric. An attempt was thus made to distinguish between gestural illustrations of verbal metaphors and gestures that were themselves metaphoric. This also explains why the gesture "came" and "went" accompanying Wang's utterance of journey metaphor, "the thunder came fast and went fast," which was discussed earlier in this section was coded as iconics rather than metaphorics.

4.4.4 Method

After transcribing the speech of the whole session, the video extract was watched by the two coders separately, focusing on just the gestures. Every gesture Wang used was categorised into one of the three types of gesture: deictic, iconic, and metaphoric. It was decided to group the gestures into just three categories rather than five (including beats and cohesives) because these three categories are more related to the focus of the study. Reasons for categorising the gestures into three rather than two (metaphorics and nonmetaphorics) were that these were the three most popular gestures and that differentiating deictic from iconic gestures would help provide a better understanding of what other types of gestures the teacher used along with metaphor in speech besides metaphoric gestures.

The 13.5 % disagreement of the coding results between the two coders was compared and discussed, until each gesture was classified into one of the three categories. The gestures were then highlighted on the transcript by using three different colours, to facilitate density and distribution analysis. Finally, functions of the gestures were noted.

4.4.5 Results

The duration of Wang's instruction in the recordings was 46 minutes, totalling 8,964 characters transcribed. In the session, 89 gestures were identified; 43% were iconics, 30% metaphorics, and 27% deictics. Every gesture was categorised. The distribution was such that over 75% of them fell in the sections on Listening to The Four Seasons (42%) and the lecture on the Baroque era (35%). Only 1% of the gestures occurred in the opening and the recorder play sections.


Wang used deictics to point at objects. Her eyebrows rose when she started the question "Can you see the word, Baroque, in the textbook" and looked at the students. Her left index finger pointed forwards. Then when Wang said the focus of the question zhege zi ("the word"), she raised her right hand and pointed at the screen behind her with her index finger. When she pronounced the word, "Baroque," she turned her upper body, half facing the screen and looked for one second at the Chinese characters for "Baroque" written on the screen.

Extract 11

[ä½  有 看到 課本] 是不是 [有 寫 這個 å­- çš„ 原æ-‡]?Baroque,有沒有?

[ni you kandao keben] shibushi [you xie zhege zi de yuanwen]? Baroque, youmeiyou?

[you have see textbook] (Q) [have write this word (DE) etymology] Baroque (Q)

Can you see if the English word, Baroque, is in the textbook?

However, deictics were not always used to point to something concrete. For example, when listening to the first movement of The Four Seasons, Wang compared the violins' trill to birds tweet in the spring. When she asked the class to pay attention to a certain part of the melody, she repeatedly put her index finger of her right hand next to her right ear and pointed to the air. Wang kept repeating this gesture whenever she tried to draw the class's attention to the music. Thus, it appeared to be the music she was pointing at, although it was not concrete, or even visible at all.

The following extract is another example of pointing to the invisible. Wang asked the class if they still remembered what instruments she had mentioned earlier in the same session. The first answer "harpsichord" came from a student and as soon as Wang heard it, she raised her right index finger, pointing. Wang then repeated the answer and gave her response, "very good." It is arguable whether Wang's finger was pointing to the word, "harpsichord," or the student who gave the answer, but in either case the gesture was categorised as deictic.

Extract 12

T: 我們 剛才 說 流行 的 樂器 有 哪些?

women gangcai shuo liuxing de yueqi you naxie

we just say popular (DE) instrument have (Q)

What are the popular instruments we just mentioned?

S: 大鍵琴




T: [] 大鍵琴,很好!還有 呢?

[] daijianqin henhao haiyou ne

[] harpsichord very good still have Q

[] Harpsichord. Very good! What else?

S: 弦 樂器

xian yueqi

string instrument

String instrument.

T: [] 弦 樂器,非常 好!

[] xian yueqi feichang hao

[] string instrument very good

[] String instrument. Very nice!


Iconics were the most common type of gesture in Wang's class, and most of them appeared in the lecture on the Baroque era and The Four Seasons section. Gestures indicating numbers and for demonstration often fall in this category. For example, Wang held both arms bent in front of her chest, with both palms facing the ground and put her tongue out, when describing a puppy sitting lazily in front of a house in summer, while listening to the second movement of The Four Seasons: "Because it's too hot, the doges put their tongues out, right?"

In the following example, Wang compared the difference between the flute and recorder while she was explaining why the recorder was translated as zhidi ("vertical flute"). A flute, however, is held horizontally by the player which is why it is also named hengdi ("horizontal flute") in Mandarin Chinese. When she asked the following question, she used both hands to imitate gestures of both flute and recorder players, to emphasise the different directions in which two musical instruments were played.

Extract 13

長笛 它 是 [ç›´è‘- 吹] é‚„ [æ©«è‘- 吹]?

changdi ta shi [zhizhe chui] hai [hengzhe chui]

flute (3SG) is [vertical blow] or [horizontal blow]

Do you play the flute vertically or horizontally?

The referent of the gestures did not always appear at the same time when the referent was uttered in speech. In Extract 14, Wang used her right hand to point to the pillars in the classroom on her right and then left side, when she was talking about the architecture of the buildings in the school. After she pointed to the pillar on her left side, she used both her thumb and index fingers of the right hand and moved the arm straight and vertically, up down and back up. These gestures were iconics, representing "straight lines" in speech.

Next, when she talked about the lines in Baroque buildings, she used her right palm to make a simple "U" curve in the air. Here in gesture, the soft "U" curve movement was contrasting with the previous gesture of moving the right thumb and index finger vertically, but in speech, the adjective "curved" contrasting with the adjective "straight" in the first line did not follow directly until the third line. The gesture preceded the word that related to it semantically. It seemed to support the hypothesis that although gestures and speech were different visual and verbal elements, somehow they were in fact conceptually integrated in an idea unit (Cienki & Müller, 2008; McNeill, 2005).

Extract 14

[柱子 是不是][都 是 ç›´ 線條 çš„]?對不對?線條 很 ç°¡å-®

[zhuzi shibushi][dou shi zhi xiantiao de] duibudui xiantiao hen kiandan

pillar (Q) all are straight line (DE) (Q) line very simple

[The pillars are][in straight lines], aren't they? Very simple lines.

可是 巴洛克 時期 的 時候 [怎麼樣?比較 繁複,對不對?]

keshi baluoke shiqi de shihou [zenmeWang bijiao fanfu duibudui]

but Baroque era (DE) time (Q) more complicated (Q)

But how about in the Baroque era? More complicated, isn't it?

[線條 比較 欸] 有 許多 的 曲折 對不對?

[xiantiao bijiao ai] you xuduo de quzhe duibudui

[line more well] have many (DE) curved (Q)

[Lines are, well,] more curved, aren't they?


In total, 30% of the gestures were metaphorics and it was interesting that metaphorics occurred in almost all the eight different classroom activities. The only two exceptions were the opening and ending remarks which Wang made; Wang used no gestures at all in her opening remarks.


TIME IS AN ENITITY MOVING TOWARD THE SPEAKER is one of the conceptual metaphors shared by both English and Mandarin Chinese. For example, in Mandarin Chinese people say shengdanjiei kuailai le, which means "Christmas is approaching," and xingqitian guo le means "Sunday passed." Time is then thereby conceptualised as something moving in space, and this can be seen even more clearly with gestures. Extract 15 was from Wang's session when she introduced the different periods in musical history. She explained the order of the Renaissance and the Baroque. Firstly she raised her left arm, straightened out her five fingers, with the palm facing down, at approximately eyebrow height, and then moved her hand down to the height of her chest. The two points in the space indicated the two different time ranges in musical history and TIME is thus represented as SPACE by the gesture. However, it is interesting in this example that time travelled in different directions in speech and gesture; in speech, time moved toward the speakers, but in the gesture it moved from up to down. In fact, the metaphor the gesture expressed here exists exclusively in terms of gesture, not in speech (c.f., section 7.4.8). That is, one would not say in Mandarin Chinese "the Baroque is at the bottom of the Renaissance" to mean the same thing.

Extract 15

æ-‡è-å¾©èˆˆ [過來 才 是 巴洛克]

wenyifuxing [guolai cai shi baluoke]

Renaissance [come yet is Baroque]

The Baroque comes after the Renaissance.


Another conceptual metaphor suggested by the gestures is SPACE (UP) AS IMPORTANCE. This has a very close relation with the common conceptual metaphor in speech: SIZE (BIG) AS IMPORTANCE. In Mandarin Chinese, da ("big") can be used as an adjective to describe something important. In Extract 16, Wang told the class that there were some important periods in musical history. When she said that "there are some important and big periods," she lifted up her left arm with her open palm facing down, then moving progressively downwards, stopped at different heights. Instead of ranking the periods from big to small by gestures, she ranked them from up to down. It was another example of different metaphors being used in speech and co-speech gestures.

Extract 16

音樂 歷史 上 çš„ 分期 å-” 有 幾個 [重要的 大的 時期]

yinyue lishi shang de fenqi o you jige [zhongyaode dade shiqi]

music history up (DE) period (PRT) have several [important big period]

About the periods in musical history, some are important and big.


Metaphoric gestures which separate different parts of an exposition appeared more than once, and although in speech Wang always said "First…moreover…and then…," she did not always use the same gestures for them. Sometimes she gestured the numbers "one," "two," and "three" even though she did not verbally say any numbers. At other times, she just turned over the other palm from facing the ground to facing upwards, when moving to a new concept or idea in speech. Such metaphoric gestures distinguish different parts of an exposition being made as separate (downwards and upwards, or right and left) spaces. For example, when Wang reviewed the main points she had mentioned in class about Vivaldi, and asked the class to write the main points down in their own textbook, she said "first…moreover…and then…," and gestured with her right index finger pointing to different fingers of her left hand, to indicate changes of topic.

4.5 Discussion and Conclusion

4.5.1 Classroom Observation

Lesson structure and activities involved

Figure 4.2 shows the relative amount of time Wang spent on different activities across the three sessions. The two main activities were recorder playing and a lecture on the Baroque era, which between them accounted for almost 2/3 of the total time. Also Wang spent 15% of time listening to The Four Seasons with the class. The two parts where metaphors and gestures were used most often were the lecture on the Baroque and listening to music, which took more than 2/5 of a session.

Figure 4.2. Pilot study: Relative time spent on activities across Wang's 3 sessions.

Gesture use and classroom atmosphere

For classroom atmosphere on the schedule, three levels were defined: high (H) was recorded when the class showed enthusiasm and the decibel level in classroom was high. Fair (F) was noted when only a part of the class responded to the teacher and the rest remained silent. Low (L) indicated that the class was quiet and seemed not to pay attention to the teacher, or when silence occurred after questions. It is important to note that the three levels were relative rather than absolute.

Figure 4.3. Pilot study: Gesture frequency and classroom atmosphere.

It was assumed that there would be a positive correlation between the number of gestures and classroom atmosphere. However, the three boxes in Figure 4.3 show the places where the two factors suggest a negative correlation. It seemed that in the parts of lecture where Wang used a certain number of gestures did not necessarily make the class more enthusiastic about their learning. One of the possible explanations was that more gestures were made due to the low classroom atmosphere.


Both the interviews I had with Wang before and after the class sessions were face to face and semistructured. Most questions I had were open questions. Making the interview semistructured allowed me to create a more natural talk atmosphere without skipping the questions to which I wanted answers. The teacher would then be free to talk whatever their feelings and/or thoughts were toward certain questions. At the same time, however, I needed to keep an eye on both time control and interview direction to make sure I got the answers to the planned questions.

Normally I did not interrupt Wang but tried to direct her back to the topics when she began to talk something unrelated. Generally speaking, Wang's answers were consistent because she echoed her own point of views when giving answers to different questions. For example, when asked how she prepared to explain new concepts (question six), she indicated that music did not exist alone and a teacher had to help students build up the whole context in terms of time, place, and how people lived and thought at that particular time, in order to understand a piece of music. She emphasised that music and other artistic forms such as architecture and arts were closely related, which she kept mentioning when answering how she decided if the materials were suitable for the class (question five) and why music education was important to her (question two).

4.5.2 Metaphoric Gestures

Functions of Metaphoric Gestures

In Wang's sessions, different functions of co-speech gestures were examined and the following were the three main ones found: (a) to emphasise, (b) to visualise, and (c) to nominate.

Firstly, gestures helped emphasise what Wang wanted to say. Usually emphasising gestures accompanied a verbal expression containing numbers, which highlighted different aspects of a topic. These gestures could be metaphorics or iconics. They were used when the teacher helped the students to either preview the main points or review the main ideas which were going to be introduced. These gestures seemed intended not only to help point out the important main points, but also to make it easier for the class to follow the teacher's exposition. Deictics which pointed out the topic being talked about in speech also seemed to emphasise points and encourage listeners to pay more attention to the topic.

Secondly, gestures were used to visualise the abstract. Wang made a big horizontal "S" shape with her left arm when telling the class that the melody they were listening to was describing the spring breeze. The metaphoric gesture suddenly made the line of the melody visible. Also, in Extract 14, Wang drew straight lines in the air while she was explaining the simplicity of the school's building compared with Baroque's complicatedness. These iconics thus helped visualise the abstract idea of simplicity.

Finally, gestures were used by Wang to express (positive) feedback. As shown in Extract 12, instead of pointing to any student after Wang asked the question to nominate a specific student to answer, she pointed after hearing the response from the students. Almost as soon as she completed the gesture, she repeated the answer from the student. It seemed that Wang's finger was activated by the voice which produced the answer, and the index finger stopped in the air and headed in the direction of where exactly the answer had come from. By doing so, the attention of the class was drawn to the answer, followed by the positive response, "very nice," made by Wang to the student concerned.

Relations of Metaphoric Gestures and Speech

As the focus is on metaphor and metaphoric gestures in this study, Table 4.1 shows the relations between utterances and co-occurring metaphoric gestures. The results support the findings from other studies in English (Cienki, 1998; Cienki & Müller, 2008).

Table 4.1

Relations Between Utterance and Co-occurring Metaphoric Gestures





The same metaphor expressed in speech and gesture

"The timbre of the piano becomes more and more delicate."

The gesture depicts a source domain ("delicate") in speech by a round and half-open palm facing up.


A metaphor expressed in gestures but not in the co-occurring speech

"First…moreover…and then…"

The index finger of one hand points to the fingers of the other hand. The metaphoric gesture distinguishes different parts of an exposition being made by representing them as separate spaces.


Different metaphors expressed in speech and gesture

"About the periods in musical history, some are important and big." (Extract 16)

Speech and gesture share the same target domain of the metaphor (important), but the source domain is characterised differently in speech (big) and the gesture (high).


Metaphors expressed by gestures never appear in linguistic form in Mandarin Chinese

"The Baroque comes after the Renaissance." (Extract 15)

What the gesture expresses here is that "Baroque is at the bottom of the Renaissance." Such an expression is not normally used in speech to mean that the Baroque comes after the Renaissance.

34.6 Implications for the Main Study

4.6.1 Classroom Observation

Researcher's role as a non participant

Although my intention was to keep the classroom as it was before my entry, this was almost impossible to do. As soon as a student noticed that there was a stranger in the room, things became different. Some students reacted to Wang's questions more actively; some tried to get Wang's attention more enthusiastically than ever, which I did not realise until Wang indicated the fact in class. Some students also kept turning around during the sessions to see what I was doing while others were curious about my reaction when a joke was told.

Wang seemed to be natural in front of the video-recorder. She seldom looked at the video-recorder and in the three sessions I observed for two days, no special reactions because of the presence of the video-recorder were found. Young mentioned that she had been observed for several times by the other teachers (pilot interview 2), and it might explain why she looked quite natural in front of it.

From the researcher's point of view, entering classrooms is a necessity for this study because it allows one to get a better idea of the classroom atmosphere and the context, which can not be completely caught by watching a videotape, let alone an audio tape. However, it can be difficult to judge if the data collected are affected due to any camera effect (Mackey & Gass, 2005). Classroom observation will remain the main method of collecting data for the main study, and further discussion on its problem and how to tackle it in the main study is given in section 5.10.

Observation schedule

The observation schedule helped me to track the process of each session during and after sessions. With the schedule, it was easy to locate where I was in the session whether I sat in the classroom or watched the recorded video at home. Overall the prepared observation schedule proved easy to mark. The categories (activity, start and end time, numbers of linguistic metaphor and gesture, participant organisation, materials, instruments, language used, and classroom atmosphere) of the observation schedule were clear, although some needed to be clarified or deleted.

It proved impossible to keep accurate tallies of linguistic metaphors and gestures during the three sessions, even though I forced myself to try to do it for a short period of time during one of the sessions. Therefore this was not used for the main study.

Some categories need to be redefined. There was a classification named "Individual" in the participant organisation section (see Appendix C) and when I designed the schedule, I was thinking about points where the teacher spent her time on just one particular student. That is, the teacher might go next to the student or ask the student to come to her, making it clear that a certain period of the teacher's time was being devoted to one single student. However, during the three sessions, this kind of individual organisation never happened. Instead, another kind of individual organisation kept appearing. Wang would talk to one particular student in front of the whole class. Although the conversation was not limited to the two (Wang and the student), it was decided to code it as "individual" participant organisation. It was decided for the main study to include both the above types of participant organisation as "individual."

Categories of pictures and audio tape confused me at first during the observation. The main material Wang used in her classes was PowerPoint slides, and the slides included not only text, but pictures and sounds. It was decided that the classification "pictures" should be extended not just to wall charts or posters, but also to pictures provided electronically. The classification "audio tape" was modified to "audio sound," which included sounds whether provided by PowerPoint or by the piano played by Wang in class. In addition, the classification "PowerPoint" was added. Hence, when Wang showed a picture of a Baroque castle via PowerPoint, both categories of "picture" and "PowerPoint" were ticked.

The category of "classroom atmosphere" was deleted for the main study because focuses of the study were modified and classroom atmosphere was no longer an issue in the main study. A modified observation schedule based on the pilot study for the main study is appended (see Appendix E).

Interview questions

After the pilot interviews, it was discovered that questions six and eight were too general for the interviewee to answer. Moreover, none of the questions was about gestures and therefore this needed to be added. Also, for questions four, five, and six, instead of asking the general situation, questions were modified to encourage the interviewee to answer based on the sessions being observed, rather than give general statements which the interviewee thought s/he "was supposed to do." Finally, the order of the questions was rearranged to create a better flow of the conversation, although it was never fixed. A list of modified questions for the main study is appended in Appendix F.

4.6.2 Gesture Coding

Gesture Coding

In this pilot, the other coder was provided a transcript with both speech and gestures. For the main study, the other coder will be provided video clips and a transcription of the teachers' speech only. S/he needed to transcribe gestures starting from identifying phases of them to avoid the subjective interpretation of my description of the gestures.

It can be very difficult to code gestures without video-recording the sessions, and even with the video, it can still take a lot of time to repeat playing one clip until gestures are properly transcribed. Both visual and audio information are crucial, for the former records the movement details without being interpreted into words, while the latter provides the speech context. Overall, gesture categories are defined not only by the hand motion but also by the role of gesture within the linguistic context. Therefore it is almost impossible to code gestures by kinetic movement data alone, and it is one of the reasons why gesture coding can be both time and effort consuming.