2.3 Limitations in Previous Studies
2.3.1 Limitations of Previous Research on the Interaction in MMORPGs
The major limitation of previous studies is that the discourse analysis conducted by the researchers might not be elaborate enough. First, some studies used an event of dialogue as the basic unit for speech act coding rather than use a complete sentence as the unit for speech act coding. The details of students’ language output might be ignored. Second, some studies did not specify their definition of certain speech acts.
The functions of some speech acts seemed identical, which makes the classification not as meaningful as expected. For example, in Reinders and Wattana’s study, the produced sentence like “What do you mean by saying that?” seemed to belong to both the category of clarification and the category of question. Third, most of the studies do not take into consideration of the percentage that each speech act accounts for in the whole interaction. Without the information, it is still hard for us to distinguish the more frequent speech acts from the less frequent ones carried out by learners, which means that the whole picture of the intercultural interaction occurring in the MMORPG worlds might be still not clear.
22
2.3.2 Limitations of Previous Studies on Incidental Vocabulary Learning in MMORPGs
As discussed in the former section, several studies have been conducted to investigate the potential of MMORPGs for vocabulary learning, but studies focusing on students’ real vocabulary gains are quite limited. There are also some limitations which need to be addressed. In Rankin et al. (2006)’s study, students’ vocabulary gains achieved statistical significance; the number of participants, nevertheless, was only four and thus the results might not be able to be generalized to other context.
Although the second study done by Rankin and his colleagues (2009) invited more students in the experiment, the researchers, however, did not take into consideration of the number of encounters with the target word in the game.
The positive effect of repeated encounters of target words has been verified by research of implicit vocabulary learning. A few decades ago, Nagy, Herman and Anderson (1985) has proposed that, in both first and second language acquisition, repeated encounters of the target word is needed to achieve full knowledge of the word because a person’s vocabulary gain is a gradual process. Repeated exposure to a word allows learners to have more exposure to its linguistic and pragmatic properties, strengthening learners’ memory of this word (Laufer & Roitblat, 2011).
Several studies have examined the number of optimal encounters of words in the context of vocabulary learning through reading. For example, Rott (1999) proposed that six encounters produced significantly larger gains. Horst and Meara’s (1998) suggested that vocabulary gains were likely to occur when words were repeated eight or more times in graded readers. Waring and Takaki (2003) discovered that it took twenty or more meetings with a word. Webb (2007) focused on the effect of word repetition on different aspects of vocabulary knowledge, including orthography, association, grammatical functions, syntax, and meaning. He discovered that as the
23
number of word presentation increased, all aspects of knowledge tended to increase.
At one encounter, participants’ improvement can be found in certain aspects. At three encounters, sizable gains were observed in every aspect; at ten encounters, significantly large learning was shown in every aspect. Although the optimal number of word occurrence needed for vocabulary learning is not yet conclusive, taken as a whole, these studies provided considerable evidence that repetition yields positive effects on incidental vocabulary learning.
In terms of former research on interaction in MMORPGs, the analysis approaches seem not elaborate enough; in terms of previous studies on vocabulary learning, they failed to take the number of participants and the importance of word repetition into consideration for vocabulary learning. The current study attempts to fill the gap and provides more insights into the affordance provides by the use of MMORPGs for SLA.
24
CHAPTER THREE METHODS
This chapter describes the methods employed in the current study. Section 3.1 provides the detailed information of the participants who take part in the study.
Section 3.2 introduces the game that adopted for research. In section 3.3, the methods of data collection and data analysis of speech acts, vocabulary test, and questionnaire will be presented. Section 3.4 contains data collection and analysis.
3.1 Participants
Ten English learners with intermediate level in English were invited to participate in this study. Their age ranges from 19 to 25, and the average age is 23.
All of the ten participants in this study are experienced MMORPG gamers with at least three years of MMORPG gameplay experience, but they have never played the game that adopted in the present study. The design that includes experienced players in the current study is to reduce the influence of participants’ novelty of MMORPG gameplay on the effectiveness of language learning. Previous studies have shown that novice players experienced difficulties in controlling avatars and utilizing in-game commands, which prevents them from engaging in successful game exploration or collaboration with other players, leading to the decrease of their motivation in using the games for language learning in the first place (Peterson, 2011;
2012). Since the whole gaming session for participants only lasts for twenty-eight days, and orientation for them is only one to two hours, the participants are expected to have already possessed sufficient gaming skills before attending this experiment.
Table 3 outlines the basic information of the ten participants, including their age, MMORPG gameplay experience, and English proficiency.
25
Table 3
Participants’ Age, MMORPG Gameplay Experience, and English Proficiency Number Participant Age MMORPG
10 Participant 10 25 12 years Intermediate
~Intermediate high TOEIC 750
3.2 The Instrument
Guild Wars 2 (GW2), a newly-released sequel to Guild War, is a well-known massively multiplayer online role playing game among game players worldwide.
GW2 immerses players in the high quality three-dimensional fantasy world, in which
26
players undertake solo play, collaborative play, duels, and match competitions and other achievement-related activities such as making handicrafts, selling and buying items of specific power, all of which make real-time conversation with other players become necessary. In-game mailbox, chat channel, and task instructions provide participants access to a great deal of English input and allow for output.
Compared with other traditional MMORPGs, GW2 is furthermore characterized by its Personal Story, which makes the game a highly potential medium for storytelling. Players are not only allowed to choose specific race, profession, and appearance but also their personality, image, dreams, or ambitions when creating their avatars. The choices players make would shape the background story of the virtual world in which the players are immersed. During the game playing process, every choice a player makes may also influence the plot or development of their own stories.
This specialty reflects Ang and Zaphiris’ (2008) statements that “computer games could be interactive narratives…the ”reader” to not only witness stories but to enact them.” Therefore, players in GW2 complete tasks with the expectation not only of character upgrading and but also of story development. Figure 1 shows the interface of GW2.
27
Figure 1: GW2 interface
3.3 Procedure
The whole procedure takes 4 weeks (28 days), with initial orientation, four-week gameplay sessions, vocabulary pretest and posttest and questionnaire administration.
In the orientation session, information about the avatars they were required to create and the timetable for task completion were provided. The participants were instructed to install the game GW2 and the video recording device on their laptops. After the device and software were installed, they began to create avatars. When creating the avatar, participants were required to play the same role, choose the same options for personal stories, and complete the same tasks in order to have the same gaming
Players can get more information about the current task by
choosing appropriate response.
Summarized instructions for the current task
Text-based chat channel
In-game mailbox
Guilds
Detailed task instruction given by NPCs
Summarized instructions for personal story
28
experience. Therefore, they would encounter the same tasks and meet the same story NPCs (detailed instructions for character creation and are shown in Appendix A).
After creating the character, the participants are required to take a pretest on target vocabulary items to assess their current knowledge of vocabulary.
During the gameplay sessions, participants were required to complete the sixteen tasks in the lowest level area, Metrica Province. In order to make sure that the participants would be able to complete all the tasks on time, the author contacted each participant regularly to check their gameplay progress. Table 4 shows the procedure of the current study.
Table 4
Procedure of the Present Study
Day Tasks Tests/ questionnaire
Before experiment
Orientation Vocabulary pretest
Day 1-7
Complte the instance in the dungeon Ascalonian Catacombs
Questionaire
In the first week (Day 1 to Day 7), the participants completed the previous eight tasks and then took the first vocabulary posttest; in the second week (Day 8 to Day 16), they completed the rest eight tasks and took the second vocabulary posttest. The 16 tasks and the sequence of these tasks which they were required to follow are shown in Table 5.
29
Table 5
The 16 Tasks and the Sequence of the Tasks Task sequence Task objective
1st Support Doola’s Golemancing research
2nd Assist Professor Gahf in teaching his young progeny 3rd Assist the Opticalium with lightning research
4th Help Blopp stop the Inquest from using hylek poison to drive lake creatures crazy
5th Use Researcher Kaii’s SCRAP-Suit to help him manage Incomp’s scrap heap
6th Assist PR&T Esoterics with ooze research 7th Help Brill Alliance study the skritt
8th Help Hrouda maintain the ecosystem in spite of Dr. Bleent’s research
9th Assist Parnna’s research into the ancient transfer chamber 10th Help Shoon’s salvage krewe deal with Inquest
11th Help the Michotl tribe
12th Help the Lionguard keep the Cuatl hylek in check 13th Assist the Chaos krewe
14th Help the C.L.E.A.N. krewe
15th Help the Luminates plant krewe maintain the power grid 16th Help Peacemaker Owta disrupt the Inquest complex
To complete each task, participants would engage in certain conversation in order to acquire the detailed information they needed for the next move. Following is one example transcript of the conversation in the task “Assist Professor Gahf in teaching his young progeny.”
Professor Gahf: Young ones are so demanding. The progeny are eager to learn,
30
but without the proper supervision, their excitement can lead to dangerous results.
Would you care to assist me? I’m in need of good assistants.
Player: What would you need me to do?
Professor Gahf: There are guest teaching apparatuses stationed around the area.
Use them to perform demonstrations for the progeny and allow them to experiment of their own. Just keep them safe from any results of their experimentation.
Player: Simple enough. What else?
Professor Gahf: Some progeny may have questions for more knowledgeable geniuses like you and me. Try to help those impressionable progeny in any way that you can.
Player: Anything else?
Professor Gahf: Conversely. There are some progeny who have chosen to ignore their studies and loaf around. Their unruly behavior must not be remain unreprimaneded.
Player: Understand. Is that all you need from me?
Professor Gahf: Be vigilant for Inquest recruiters. Progeny are more susceptible to their deception. Chase them off through intimidation, or physical force if necessary.
But we mustn’t allow them to contact with the progeny.
Player: Got it. I’ll get to work.
In the third week, where the participants had already finished the 16 tasks and their avatar had reached more than level 15, they could explore the virtual world based on their own will. They were required to achieve level 30 before day 21, because level 30 in GW2 is the threshold for joining a group to accomplish the major task on the special map, which is called “instance.” During the gameplay, participants are not allowed to look unknown words up in the dictionary and thus the number of word encounter becomes the sole variable of vocabulary learning in the research.
In the fourth week, the participants (with avatars at level 30 or above) were required to choose a time period to join a group play and complete the first instance of the game, which occurs in the dungeon called Ascalonian Catacombs (AC). They could either find their group members in the guild that they attended or just invite
31
other players near the dungeon. They were required to find at least three or four other members to join their group because the teamwork of four or five people is necessary for group play in AC. They need to speak in English to invite others and make sure that the players whom they invited speak English to them. Since the game comes in English and Europe servers only, most of the players are from English-speaking backgrounds. It is assumed that the team members whom the participants found were native speakers of English or at least players with comparatively higher English ability. In the present study, they would be termed as “fluent English speakers.” The duration of the instance is about one to two hours, depending on members’ combat effectiveness and collaboration.
3.4 Data Collection and Analysis 3.4.1 Video Recording
To answer research question one that what kind of speech acts that language learners carry out during the interaction in the MMORPG world, the ten participants’
gameplay was recorded when they participated in the instance with other random players in the dungeon Ascalonian Catacombs, where they were expected to have more interaction with other players who have better command of English. The game recording software, Bandicam, was used in this study to record their text messages occurring in the in-game instance message channel or in the Skype. The collected data would be analyzed on the basis of the framework of speech acts the author designed, as shown in Table 6.
The framework was devised on the basis of Searle (1976)’s and Fraser (1975)’s, taxonomy of speech acts, integrating Long (1983)’s concept of social interaction.
Each category in the framework contains several subcategories with reference to the findings of previous studies (e.g., Peterson, 2012; Reinders & Wattana, 2011; Zeng et
32
al, 2012) and the author’s data in pilot study on one player’ produced speech in the gaming process. There are seven major types of speech acts in the taxonomy, including representatives, directives, expressives, declarations, negotiation of meaning, greetings and farewells, language specific correction.
The first one is Representatives. Speech acts pertaining to this category function to tell people how things are and to suggest various degrees to the truth such as asserting, claiming, reporting or doubting something (Searle, 1976). Several subcategories are identified, including providing factual information, providing perspective or opinions, seeking factual information, seeking perspective or opinions, expressing need, reporting on actions, affirmation, and locating.
The second category is Directives. In this class, the primary point of the speech acts is to try to make other people do something, including speech acts such as commanding, requesting, and suggesting (Searle, 1976). This class contains some subcategories, including making requests, giving commands, and inviting others.
The third category is Expressives. The function of this class is to express the speaker’s attitudes or feelings about certain thing such as apologizing, regretting, or showing gratitude (Searle, 1976). There are several subcategories, including showing gratitude, apologizing, expressing excitement and emotion.
The function of speech acts in the class of Declarations is to make changes in the world simply through their successful realization such as the sentence “you are fired,”
said by a boss or “I do,” said by the bride (Searle, 1976). Its subcategories are accepting and rejecting.
The speech acts of Negotiation of meaning were frequently realized in daily conversation. According to Long (1983), learners negotiate with their conversation partners in order to solve comprehension difficulties so that incomprehensible input may become comprehensible, which leads to language acquisition. Negotiation is
33
usually achieved by checking for confirmation, checking for comprehension, and requesting clarification. This class, therefore, contains confirmation checks, comprehension checks, and clarification requests.
Greetings and farewells are also an important speech acts in conversation, including greetings and farewells. As for the category of language specific correction, it involves correcting others’ language and correcting their own language
There are 22 speech acts in total in the taxonomy used in the present research. The transcript of students’ produced speech will be analyzed. The classification of speech acts and examples are presented in Table 6.
Table 6
Classification of Speech Acts and Examples in the Current Study Categories of
speech acts
Coding number
Subcategories examples
Representatives s1 Providing factual information
The fiend is coming in this route.
s2 Providing
perspective/opinions
This character is hard to use.
s3 Seeking factual
s5 Expressing need We need one more person to join us.
s6 Reporting on actions I went out to find some help
s7 Showing being aware
of something
A: I’m leaving.
B: Ok./Alright.
s8 Locating The dungeon Ascalonian
Catacombs
Directives s9 Making requests Could you lead me to the waypoint?
34
s10 Giving commands Go to the top now.
s11 Inviting others Do you want to go to AC with us?
Expressives s12 Showing gratitude Thanks.
s13 Apologizing Sorry…
s17 Comprehension checks Do you understand my questions?
s18 Clarification requests What do you mean by saying that?
A: Actually I kinda heye that dwarf.
A: hate
3.4.2 Vocabulary Test
In response to research questions two, 54 words which participants would encounter in the gaming process were identified for testing (as shown in Appendix B).
Because the whole gaming process was long, so the posttest to assess vocabulary learning was divided into the 1st posttest and the 2nd posttest. In sum, there were three sets of tests in total, including the pretest, 1st posttest, and 2nd posttest.
35
In the pretest, there were 54 target vocabulary items which participants might encounter from task one to task sixteen. The 1st posttest contained the former half vocabulary items selected from task one to task eight (24 words) and the 2nd posttest contained the remaining vocabulary items from task nine to task sixteen (30 words).
The selection of the target vocabulary items was based on the following criteria:
a. The difficulty of the word is at college level (but context specific words such as “moa” or “golem” are excluded).
b. The word occurs in the main stories, events, and task instruction.
c. The word appears at least five times during gameplay, which allows participants to have more exposure to this word.
Participants needed to write down the meaning of each word either in Chinese or in English. The part of speech of words was given in order to avoid participants’
confusion of word meaning (see Appendix C). 1.85 points were given to each correct answer. An example is provided below:
progeny (n.)
Do you know the meaning of this word?
□ No.
□ Yes. It means ______________________________
The mean differences between pretest and posttests were analyzed quantitatively through Paired-Sample T-test of SPSS 18.0 to examine if there was significant difference.
3.4.3 Questionnaire
3.4.3 Questionnaire