---
title: 論文研討
tags: 碩一, 作業
---

# 問題導向的遊戲和學習教的交互作用在學生的英語聽力表現和行為模式
Interaction of problem-based gaming and learning anxiety in language students' English listening performance and progressive behavioral patterns

## 摘要
- 學者已經指出將學習內容整合進遊戲的重要性，但很少深入探究在遊戲學習環境中，學習者的學習過程
Scholars have indicated the importance of considering anxiety in language learning. They have also pointed out the potential of integrating learning content into gaming contexts. However, few have explored in-depth the learning processes in game-based learning environments while also taking students' learning anxiety into account.
- 此次研究開發「問題導向的英語聽力遊戲」，以77名九年級生（國三生）進行準實驗以評估學生的「學習成績、學習動機、英語焦慮」
To address this issue, a problem-based English listening game was developed, and 77 ninth graders took part in the study by playing the game. A quasi-experiment was conducted to evaluate students' learning achievement, learning motivation and English anxiety.
- 此外，使用「progressive sequential analysis」分析不同英文焦慮程度的學生，他們的學習行為模式的不同
In addition, progressive sequential analysis was employed to explore the learning behavioral patterns of students with different levels of English anxiety.
---
- 發現遊戲方法有利於學生的學習成績和動機
According to the three learning phases, it was found that the gaming approach benefited the students' learning achievement and motivation.
- 英文焦慮較高的學生會有比較繁雜的學習方式與遊戲行為，而且會有比較好的學習表現，比起英文焦慮較低的學生
Moreover, the students with higher levels of English anxiety progressively performed more complex learning and gaming behaviors in the gaming context, and had better learning achievement than those with lower levels of anxiety.

## 簡介
- 近年來研究指出，「學習動機、學習滿意度（learning motivation and satisfaction, ）」可以帶給學生在學習上正面的影響
- 「學習焦慮與分心（learning anxiety and distraction）」則可能導致學習效果不佳，尤其是語言學習，這種需要在相關環境下反覆練習的學習
In recent years, the relationship between students' affect and their learning performance has been considered as an important issue in education (Cheng, Su, Huang, & Chen, 2014; Verkijika & De Wet, 2015). Several studies have reported that students' affect is associated with their cognitive learning (Konak, Clark, & Nasereddin, 2014; Valentín et al., 2013). Moreover, positive affect, such as learning motivation and satisfaction, can encourage students’ engagement in learning (Cheng et al., 2014; Wong, Chai, Aw, & King, 2015), while negative affect, such as learning anxiety and distraction, could lead to poor learning outcomes, in particular in language learning which requires frequent practice in relevant contexts (Kim & Glassman, 2013; Teo, Tan, Yan, Teo, & Yeo, 2014).
- 學習外語關係到學生的「學習動機與外語的學習焦慮」，尤其對外語的學習焦慮可能成為學生的學習成績決定因素
It is recognized that learning a foreign language is strongly associated with students' learning motivation and their foreign language anxiety defined foreign language anxiety as a feeling of worry, nervousness, and apprehension when learning a foreign language. Previous researchers have indicated that foreign language anxiety can be a determinant of students' foreign language learning achievement.
- 2016年學者發現對外語的焦慮影響伊朗學生用英語交流的意願，2015年學者證實對外語的焦慮會影響學生外語溝通能力，因此有很多學者常使各種學習策略減少學生的外語焦慮，其中一種是透過遊戲式學習方法（game-based learning）
For instance, Khajavy, Ghonsooly, Fatemi, and Choi (2016) examined Iranian students' willingness to communicate in English, and found that the students' perceived competence of English and their English anxiety directly affected their willingness to communicate in English. On the other hand, Denies, Janssen, and Yashima (2015) also proved that their level of foreign language anxiety plays an important role in students' communication in foreign languages. Consequently, an increasing number of researchers have attempted to reduce students' anxiety while learning English by integrating several learning strategies and technologies, especially employing game-based learning approaches.
- 學者認為遊戲式學習可以有效降低學生的學習焦慮，因為遊戲的娛樂性、互動性、即時動態反應等等性質。例如：GeCALL
Some researchers have found the game-based learning strategy to be a favorable learning approach for reducing students' learning anxiety because of the enjoyment, interactivity, and dynamic nature of the games.
For instance, ==Young and Wang developed the Game Embedded CALL (GeCALL) system for students' English pronunciation practice.== The results revealed that this learning approach can reduce students' English speaking anxiety and provide them with more opportunities to do speaking practice
- 雖然很多人探討遊戲式學習對學生學習焦慮的影響，但很少人對學生的學習行為作分析，而更好的研究其實是分析學習行為，並且釐清不同學習焦慮的人，他們的學習行為的不同。
Although many studies have discussed the treatment of game-based learning for students' English anxiety, few have explored the relationship between students' actual learning behaviors and their English learning anxiety. Since the analysis of learning behaviors has been considered as an approach to understanding more precisely how students behave, it would be better that researchers explore students' learning behaviors and clarify the different learning behaviors of students with varying levels of learning performance.
- 基於上述所言，此研究開發「問題導向的英文聽力學習遊戲」以探討學生在每個遊戲階段進步的學習行為，與不同英文焦慮的學生，他們在學習行為上的不同
In order to explore the role of English anxiety and game-based learning strategies in students' English learning, in this study we ==developed a problem-based English listening game== to help students with their English learning. Moreover, analysis of the students' progressive behavioral learning patterns was employed to probe their learning behaviors in each gaming phase. Finally, an exploration of the students’ progressive behavioral learning patterns according to the level of their English anxiety was conducted.

## 文獻探討
### Issues related to English anxiety
#### English anxiety 英文焦慮
- 英文焦慮的定義
The issue of performance associated with anxiety in English learning has been documented, and researchers have found that most students who are weak in English have difficulty speaking and discriminating the sounds and structures of a target language message (Yan&Horwitz, 2008). English anxiety has been conceived as“ the feeling of tension and apprehension specifically associated with second language contexts, including speaking, listening, and learning”(MacIntyre&Gardner,1994).
> 多數英文較弱的學習者是因為難以說出或判斷英文的語言結構與發音差異。 [name=Yan&Horwitz, 2008]
> 英文焦慮定義為「對於第二外語相關事物，包括口說、聽力與其他相關學習感到焦慮、緊張」 [name=MacIntyre&Gardner,1994]
- 英文焦慮的學生通常會失去信心，放棄練習或學習外語，甚至出現迴避等負面心態。
Generally, students tend to give up practicing English and lose confidence when they experience higher English anxiety. For instance, anxious students may exhibit avoidance behaviors, have difficulty learning vocabulary by heart, and have negative attitudes towards answering questions in class.
Previous findings (Alipanahi&Mahmoodi, 2015; Lin, Chao,&Huang, 2015) have clearly indicated a negative relationship between English anxiety and English performance. When students' English anxiety increases, their English learning performance decreases. A number of studies have sought and clarified determinants associated with English anxiety (Khajavyet al., 2016; Tum, 2015). For instance, Zhang (2013) examined the relationship between students' English listening anxiety and their English listening performance, and found that the students' English listening anxiety could affect their English listening performance. Khodadady and Khajavy (2013) further indicated that external motivation and being less self-determined are positively related to students' English anxiety. In addition, Peng (2015) suggested several factors associated with students' English anxiety, including English-specific images of their desired future, the English-specific attributes that they believe they ought to possess, and their previous experience of learning English.
> 英文焦慮程度和英文表現成反比關係，當學習者的英文焦慮增加時，因文表現反而會下降 [name=Alipanahi&Mahmoodi, 2015; Lin, Chao,&Huang, 2015]
> 有學者研究英文聽力焦慮與學習表現，發現英文聽力焦慮會影響學習者的學習表現 [name=Zhang (2013)]
> 學者指出「外部動機、[self-determined](http://terms.naer.edu.tw/detail/1305152/)」與學習者的英文焦慮成正相關 [name=Khodadady and Khajavy (2013)]
> 學習者對與英文相關未來的想像，和他們相信他們應具有的英文相關能力，以及他們先前學習英文的經歷都會影響他們的英文焦慮 [name=Peng (2015)]
- 英文焦慮會影響，非以此為母語的學生的英文學習表現
According to the literature, it was concluded that English anxiety would affect non-native students' behaviors and English learning performance (Yen, Hou,&Chang, 2015). For instance, in Taiwan, English is regarded as an important language and a necessary competence. However, students face some challenges such as lacking a practical environment for practicing and applying English, limited learning time in school, and being passive when it comes to speaking English openly around others because they are afraid of being despised (Yang&Chang, 2008). Therefore, the issues of reducing students' English anxiety and decreasing teachers' difficulties in teaching English have become important in Taiwan (Young&Wang, 2014).
> 英文焦慮會影響，非本地的學生的英文學習表現 [name=Yen, Hou,&Chang, 2015]
> 例如台灣，雖然英文被認為是重要的外語且必要的能力，但學生卻必須面對缺乏練習與應用的環境，僅僅在學校學習的有限時間。因此，每當學生必須在公開場合以英文溝通，就會變得被動，因為他們擔心自己不足的練習與實力會遭到他人的鄙夷 [name=Yang&Chang, 2008]
> 在台灣，減少學生的英文焦慮、減輕老師英文教學的困難，是一個越來越重要的課題 [name=Young&Wang, 2014]

### Related computer-assisted English learning
- 在英文教育上新技術的發展
Studies focused on integrating new strategies and technologies into learning have been fruitful in the past decades (Clark&Mayer, 2012, pp. 368e 398;Kozma, 1991). Various educational applications have been developed and employed in regular English learning courses. For instance, Hsu (2015) developed a video-based English learning system, and provided students with adaptive assistance based on students' learning needs. The results indicated that the students who learned with this approach practiced English more; moreover, their intrinsic motivation also improved more than that of those who learned with conventional video watching. Another example is Yang, Gamble, Hung, and Lin's (2014) investigation of the effectiveness of adaptive English literacy instruction for students' acquisition of critical thinking skills and English literacy. The findings of their study demonstrated that the students were more active in their online discussion, and that their critical thinking skills as well as their English literacy improved.
> Hsu 開發一套英文影片學習系統，證明此種系統提供學生更多的英文練習，且內在動機會比觀看傳統影片的學生要來得高 [name=Hsu (2015)]
> 其他例子為，新技術的教學模式可以提高學生在線討論度，與增加學生批判式思考和英文素養 [name=Yang, Gamble, Hung, and Lin’s (2014)]
- 新技術的發展提供學生更多練習機會，並且減少他們英文焦慮程度
On the other hand, many researchers have indicated that technology plays an important role in providing students with more opportunities to practice their English-related competence and reduce their English anxiety (Yen et al., 2015). For instance, Sun and Yang (2015) integrated service learning into an English as a Foreign Language (EFL) speaking class, and employed Web 2.0 tools to assist students in practicing English. In the learning activities, students had to record their pre-sentations in English and share their videos on Facebook, where they could discuss them with their peers. The experimental results indicated that this integration enhanced the students' speaking skills and improved their confidence in speaking English. In another study, Hung (2015) conducted flipped-based WebQuest active learning in an English class, and found that the learning approach could assist students in obtaining positive learning attitudes, and encouraged them to devote more effort to their learning. Moreover, the students' English learning achievement improved. The above studies focused on developing and adopting educational applications to facilitate students' English learning.The findings of these studies also indicated the positive support of integrating technologies into English learning, as they provided students with more opportunities to use English, and benefited their learning performance while also reducing their English anxiety.
> Sun and Yang 整合服務學習與EFL(英文作為第二外語)的口說課程，使用 [Web 2.0](https://zh.wikipedia.org/wiki/Web_2.0) 的技術輔助學生。此課程，學生必須事先在家錄製英文口說影片，並且上傳至 facebook 與其他同儕討論。結果表明，這項研究不但提高學生的口說能力，也提高學生的信心。 [name=Sun and Yang (2015)]
> Hung 使用了翻轉教室的原理，在英文課程上，採用了 WebQuest active learning 的學習活動。這項活動讓學生以更加積極的學習態度面對英文，且鼓勵他們努力研讀英文。對於英文的表現也有提升。 [name=Hung (2015)]
- 上述的實驗都有對學習者產生正向的幫助，且都有助於減少學生的英文焦慮
### The potential of game-based English learning environments
- 遊戲式學習的快速發展
Recently, the investigation of game-based learning has developed rapidly (Dickey, 2006; Griffiths&Davies, 2002), and this learning approach has become popular for increasing students' engagement in learning and improving their learning motivation (Hwang, Chiu,&Chen, 2015). In 2001, Prensky indicated that learning in game-based environments could address some problems faced in traditional instruction by allowing students to learn at their own pace and receive some adaptive learning content (Prensky, 2001). Moreover, from the constructivist perspective, the gaming strategies for learning can encourage students to pay more attention to the learning content while they are playing (Tsai, Yu,&Hsiao, 2012)Furthermore, Kiili (2007) suggested that educational games need to focus on leading students to reflect on the game-based learning context, indicating that students review the learning content and make reflections when they fail to solve tasks, and then change their strategies accordingly to solve the tasks they face.
> 遊戲式學習正迅速發展 [name=Dickey, 2006; Griffiths&Davies, 2002]
> 遊戲式學習增加學習者學習的樂趣，且提升他們的動機，因此蔚為流行 [name=Hwang, Chiu,&Chen, 2015]
> 遊戲式學習藉由讓學生以自己的腳步學習與適應學習內容，解決傳統建構式學習所面臨的問題。 [name=Prensky, 2001]
> 以[建構主義](https://zh.wikipedia.org/wiki/%E5%BB%BA%E6%A7%8B%E4%B8%BB%E7%BE%A9_(%E5%AD%B8%E7%BF%92%E7%90%86%E8%AB%96))來看，遊戲學習策略鼓勵學生在玩遊戲之餘更專注在學習內容上 [name=Tsai, Yu,&Hsiao, 2012]
> 遊戲式學習應該專注於引導學生在遊戲中的反應，在學生於遊戲中任務失敗時，指導學生複習學習內容。依據他們解決任務的情況改變策略。 [name=Kiili (2007)]
- 遊戲式學習可能產生的負面影響
Nevertheless, several scholars have noted possible negative effects of conducting game-based learning activities, such as the risk raised because of the lavish entertainment factor of educational games (Kim, Park,&Baek, 2009). While students' curiosity and attention may be aroused, learning content could easily be ignored; thus, they might only gain limited knowledge during the gaming process. This suggests the need to integrate instructional strategies into educational games or gaming activities to guide students to accomplish learning tasks rather thanmerely adopting commercial games that might not completely meet the educational objectives (Erhel&Jamet, 2013).
> 由於遊戲的特性，奢侈的娛樂性，可能會提高對學習導致負面影響的風險 [name=Kim, Park,&Baek, 2009]
> 學生可能對於遊戲挑起興趣和好奇心，但學習卻被忽略，因而僅學習到有限的知識。這意味著教學策略必須整合到教育遊戲裡，或透過遊戲活動指導學生完成任務，而非僅僅使用無法達到教學目的之商業遊戲。 [name=Erhel&Jamet, 2013]
- 英文教育與遊戲式學習
In the English learning field, some researchers have found that the game-based learning strategy benefits students' English learning (Reinders&Wattana, 2014; Sandberg, Maris,&Hoogendoorn, 2014). For example, Hung, Young, and Lin (2015) developed a collaborative and competitive game-based learning environment to improve students' English proficiency. According to both the quantitative and qualitative results, the study found that the proposed learning environment can lead students to have better learning interaction with their peers, while also reducing the achievement gap for disadvantaged students. Another example is Liu and Chu's (2010) study in which they conducted an English course in a context-aware ubiquitous game learning environment where students could practice listening and speaking related to the learning zone (Campus Life). It was found that the students' learning achievement and learning motivation improved, and the relationship between the two variables presented a positive correlation.
> 一些研究學者發現遊戲式學習有助於英文學習 [name=Reinders&Wattana, 2014; Sandberg, Maris,&Hoogendoorn, 2014]
> Hung, Young, and Lin 開發一款競爭與合作的遊戲學習環境以改善學生英文水平。結果顯示，這種環境有助於學生與同儕互動，且減少弱勢學生的成績差距 [name=Hung, Young, and Lin (2015)]
> Liu and Chu 他們在 [context-aware ubiquitous](http://newsletter.teldap.tw/news/InsightReportContent.php?nid=4395&lid=498) game learning environment 進行英文課程，學生可以練習與學習地域相關（例如學校生活）的口說練習。他們發現學生的學習成績與學習動機都有所成長，且兩變量呈正相關。 [name=Liu and Chu (2010)]
- 一些學者聲稱遊戲式學習可以減輕英文焦慮並且促進英文學習
In the meantime, many studies have claimed that game-based learning strategies are beneficial in terms of alleviating students' English anxiety as well as facilitating their English learning. They have also indicated that students' affect, such as English anxiety, is associated with their learning performance (Konak et al., 2014). For instance, Young and Wang (2014) demonstrated a game-based learning environment that engaged students in a “stress-free” context to practice their English speaking skills, while Reinders and Wattana (2014) conducted the “Ragnarok Online” investigation into students' English learning courses and found a significant improvement in the students' willingness to communicate as well as a significant decrease in their anxiety when communicating in English.
> Reinders and Wattana 使用「Ragnarok Online」的線上活動，發現此活動改善學生溝通的意願，並且減輕他們用英文溝通的焦慮 [name=Reinders and Wattana (2014)]
---
- 但很少探討關於在遊戲學習環境中學習行為與英文焦慮的變化，以及不同英文焦慮程度學生的行為模式的不同。這限制了我們了解弱勢學生在學習過程中所面臨的困境。
On the other hand, discussion has seldom focused on the in fluence of English anxiety on students' behavior in game-based learning environments. Besides, few studies have investigated the behavior in gaming contexts of students with different levels of English anxiety, which may limit educators' understanding of the learning process that disadvantaged students face.
- 因此本研究在探討學生在遊戲中的學習行為，此外探討不同英文焦慮程度的學生他們的學習行為差異
Therefore, in this study, an exploration of students' learning behaviors in a game-based environment was conducted. Furthermore, the differences in the learning behaviors of students with varying levels of English anxiety are discussed.
- 本研究共有兩個步驟，第一，開發一款「問題導向的英文聽力遊戲」且設計準實驗，以確認遊戲實際上有助於學生的學習成效；第二，探討不同英文焦慮程度學生的行為。
Accordingly, two research steps were included in this study. First, a problem-based English listening game was developed and a quasi-experimental design was conducted to ensure that the proposed game did in fact benefit students' learning achievement. Second, an exploration of the behaviors of students with different levels of English anxiety (higher and lower English anxiety) was conducted.
- 研究問題如下：
1. 問題導向的英文聽力學習遊戲，比起傳統的應用科技技術的教學，有沒有對於學生的學習成效有所幫助。
2. 問題導向的英文聽力學習遊戲，比起傳統的應用科技技術的教學，有沒有增加學生的學習動機。
3. 問題導向的英文聽力學習遊戲，比起傳統的應用科技技術的教學，有沒有減少學生的英文焦慮
4. 高英文焦慮的學生和低英文焦慮的學生，在學習模式上是否有不同之處
5. 高英文焦慮的學生和低英文焦慮的學生，在學習成效上是否有不同之處

The research questions addressed in this study are as follows:
1. Does the problem-based English listening game benefit the students' learning achievement more than the conventional technology-enhanced instruction?
2. Can the problem-based English listening game enhance the students' learning motivation in comparison with the conventional technology-enhanced instruction?
3. Can the problem-based English listening game reduce the students' English anxiety in comparison with the conventional technology-enhanced instruction?
4. Are there differences between the learning behaviors of the students with lower and higher levels of English anxiety learning when playing the problem-based English listening game?
5. Are there differences between the learning achievements of the students with lower and higher levels of English anxiety learning when playing the problem-based English listening game?

## 問題導向的英文聽力遊戲的架構與開發
- 選擇英文聽力作為英文學習範疇的教學範圍
In order to explore the role of English anxiety in students' game-based English learning, we first chose the learning content in the current educational game, that is, practicing students' English listening skills. Previous researchers have proposed that English anxiety is usually raised because of facing the challenge of English listening (Yan&Horwitz, 2008); however, English listening skills are regarded as a basic competence that Taiwanese students need to be proficient in (Chou, 2015; Yang,Chuang, Li,&Tseng, 2013).
> 英文焦慮通常會隨著面對英文聽力的挑戰而提高 [name=Yan&Horwitz, 2008]
> 在台灣公認英文聽力是台灣學生必須精通的基本能力 [name=Chou, 2015; Yang,Chuang, Li,&Tseng, 2013]

- 這項研究開發一套「問題導向的英文聽力遊戲」
- 以 RPG Maker 作為開發工具。
- 遊戲的組成資料
- 學習資源資料庫：由一位教書十多年的英文老師提供，包括英文聽力任務、生字字卡的聽力素材、英文聽力的指導
- 遊戲元素資料庫：包含戰鬥、獎勵、寶藏、武器等遊戲元素
- 學生紀錄資料庫：記錄學生學習狀態和在遊戲中的學習行為

Accordingly, in this study, a problem-based English listening game was developed to assist students in developing and practicing their English listening abilities. Fig. 1 shows the system structure of the problem-based English listening game, in which RPG Maker was adopted as the gaming tool to assist students' English listening practice. The game was composed of a learning materials database, a gaming elements database, and a students' profile database. The learning materials in the game were provided by an experienced teacher who has taught English for over ten years. The learning materials database presents some English listening tasks, a dictionary, vocabulary listening cards, and some instruction on English listening skills. On the other hand, the gaming materials database includes some battles, rewards, treasure and weapons. Finally, the students' profile database automatically recorded the students' learning status and their learning behaviors in the game.
- 遊戲有三個階段，每個階段包含一個遊戲目標和一個學習目標
There are three phases of the game that the students have to play, each including one learning goal and one gaming goal, as shown in Fig. 2. The difficulties of the goals from phases 1 to 3 are low, medium, and high, respectively. For instance, the learning goal of phase 1 is picture identification tasks which require students to look at the picture and listen to the options and then choose the correct answer. The second phase is listening and short answer questions which focus on the students' listening and responding with the correct answers. The last phase is listening and comprehension tasks which require the students to listen to conversations and infer their meaning.
Several learning tasks and gaming tasks were included in each phase of the game, which requires students to solve all of the tasks in order to achieve their goals in each phase. Fig. 3 shows examples of the learning tasks in each phase. The students could solve those tasks by employing the learning materials they had learned before performing the learning tasks. When they learned from those materials and completed the learning tasks, some of their virtual abilities would improve, such as their battle skills and the power of their weapons. Therefore, the students would have the ability to complete their gaming tasks.
- 在遊戲一開始會介紹遊戲、故事背景、規則。這個遊戲從一所巫師學院開始，學生必須經歷一連串的挑戰，最後從巫師學院畢業。遊戲過程中，學生會和 NPC 互動，完成任務，學生必須增進他們的能力已完成特定困難的任務。有些小關卡或練習會在學生接受任務之前出現，訓練學生下一關的能力。另外有一些遊戲元素，包括任務成功會有小金幣或鑰匙等等獎勵元素，當他們探索環境的時後可能會或得工具或武器，鼓勵學生加入戰鬥或解決遊戲任務等等，投入遊戲與學習。
At the beginning of the game, an introduction to the background story, the rules and the final goal of the game are presented. The storyline of this game is related to a challenge for graduating from a school of wizardry. The main character of the game is an undergraduate student who has to pass the challenges in order to graduate from the school. Therefore, she has to complete the learning tasks in each gaming phase. Fig. 4 shows the interface of the game, where students interact with the NPCs and choose the tasks they are going to complete, including the learning tasks and gaming tasks. They need to improve their skills by completing specific learning tasks, or their skills would not be good enough for them to solve the related gaming tasks.
Some learning materials are provided for the students to practice their listening skills before they accept the learning tasks. For instance, they can hear the pronunciation of the vocabulary from the vocabulary cards, look up some higher difficulty level sentences in the dictionary, or learn some tips about English listening skills from some NPCs. In addition, some gaming elements are included in the game. As students complete the learning tasks, some corresponding awards would be given in the game. For instance, when students look up the dictionary in the library, their experience points increase; some coins or keys are given as soon as they finish their learning tasks; moreover, theyget some weapons or tools as bonuses during their exploration in the gaming context. These gaming elements are important for encouraging the students to join the battle and solve the gaming tasks.
## 實驗設計
- 此研究目的是為了探究藉由玩遊戲的方式學習，不同英文焦慮程度的學生他們學習行為。研究一開始開發一套「問題導向的英文聽力遊戲」，並且進行準實驗，確認遊戲有助於學生。然後比較基於此遊戲學習的學生與傳統教學的學生。之後研究英文焦慮對學習行為的影響。此研究藉由遞增分析，分析不同英文焦慮的學生在遊戲中的時間來推斷學生的學習行為。
The aim of this study was to explore the learning behaviors of students with different levels of English anxiety while learning by playing games. This study first developed a problem-based English listening game and conducted a quasi-experiment to ensure the educational game was beneficial for the students. Therefore, a comparison of the students who learned with the problem-based English listening game and those who learned with conventional instruction was conducted.
After that, an exploration of the role of English anxiety in the experimental students' learning behaviors was conducted. This study employed a progressive sequential analysis to infer the learning behaviors of students with different levels of English anxiety by considering the time they were immersed in the problem-based English listening game.
### 參與者
- 一共 77 名九年級生（國三生）參與實驗，在實驗之前這兩個班皆為同一個老師進行一般的常規教育，受試者的平均年齡為 15 歲。
- 一個班 39 名學生，為實驗組，透過問題導向英文聽力遊戲，來練習英文
- 一個班 38 名學生，為控制組，進行傳統教學，透過投影片上課來學習，然後透過普通的音樂播放器來練習聽力技巧。
- 為了保護受試者，實驗按照台灣授權倫理委員會建議的道德標準進行，且經由受試者家長同意，並且保護受試者個人資訊，且受試者皆自願參與，若不願意，隨時可退出實驗。
There were, in total, 77 ninth graders who participated in this study. One class, including 39 students, was assigned to be the experimental group and learned with the problem-based English listening game. The other class (38 students) learned with the conventional instruction, that is, the teachers taught the same learning content with slides, and the students practiced their English listening skills by listening to the music player. The two classes were taught by the same teacher in their regular English courses. The age of the participants was 15 on average. To protect the participants, the experiment was conducted following the ethics criteria suggested by an authorized ethics committee in Taiwan. That is, participation of the children was approved by their parents, and the participants were protected by hiding their personal information during the research process; moreover, they knew that their participation was voluntary and that they could withdraw from the study at any time.
### 工具
- 前測與後測
- 通用 GEPT（全民英檢） 作為標準
- 題型：聽然後選圖片、聽然後選適合的回答、聽完對話然後選正確的答案
- 然後兩個測驗題目沒有相關聯，並且難度相當
> 台灣九年級生的水平大約是初級或中級的水平 [name=Roever&Pan, 2008]
- 學習動機問卷
- 改編自 Pintrich and De Groot(1990)
- 七個項目五點量表
- 英文焦慮問卷
- 改編自 Horwitz, Horwitz, and Cope (1986)
- 三十二項目五點量表

The instruments adopted in this study were the pre-test, the post-test and the questionnaires of learning motivation and English anxiety for measuring the students' perceptions of the learning activity. The difficulty levels of the pre-test and the post-test were determined by the General English Proficiency Test (GEPT). The GEPT is a test commissioned by Taiwan's Ministry of Education which evaluates students' English performance at five levels: Elementary, Intermediate, High-Intermediate, Advanced, and Superior. Moreover, a previous study revealed that the average levels of ninth graders in Taiwan are Elementary and Intermediate levels (Roever&Pan, 2008). Therefore, another two English teaching experts determined and selected the items from the Elementary and Intermediate levels, and then randomly assigned them to the pre-test and post-test. Both the pre-test and the post-test examined the students' performance of English listening using multiple-choice items.
Furthermore, both tests evaluated the students' English listening skills from three kinds of questions: look at the picture and choose the answer, listen and choose the proper response, and listen and analyze the meaning of the conversation. There were 3, 7, and 10 items respectively for each kind of question in each test. The Pearson's correlation between the two tests was 0.65 (p < 0.001). This high and positive correlation means that the two forms were parallel, indicating that the level of difficulty of the two tests was the same.
The learning motivation questionnaire was modified based on the questionnaire developed by Pintrich and De Groot(1990). It consisted of seven items with a five-point rating scheme (from 1-strongly disagree to 5-strongly agree), such as “I think that what I am learning in this class is useful for me” and “I like what I am learning in this class.” The Cronbach's a value of the questionnaire in this study was 0.92, showing acceptable reliability in internal consistency.
On the other hand, the English anxiety questionnaire was modified from the Foreign Language Classroom Anxiety Scale proposed by Horwitz, Horwitz, and Cope (1986). It consisted of 32 items with a five-point rating scheme (from 1 - strongly disagree to 5 - strongly agree), such as “I never feel quite sure of myself when I am speaking in my foreign language class”and “I keep thinking that the other students are better at languages than I am.” The Cronbach's a value of the questionnaire in this study was 0.94, showing acceptable reliability in internal consistency.
### 行為代碼編寫
- 兩個階段
- 第一個階段
- 玩英文聽力遊戲可能經歷十種行為，大約可分為兩類：學習行為和遊戲行為
- 確認行為代碼，是否符合研究目標
- 第二個階段
- 將行為代碼嵌入遊戲中，例如學生打開書櫃出現一個短語，為行為代碼中的「R_P（閱讀短語）」，將代碼寫入資料庫以方便事後研究學生的行為

In order to explore the students' learning behaviors in the problem-based English listening game, a coding scheme was developed to code the students' behaviors. To ensure the reliability of the behavioral analysis in this study, three researchers participated in the two-stage coding process. In the first stage, the researchers defined ten behaviors that students were likely to experience in the problem-based English listening game by referring to the literature (Hou, 2012; Lai&Hwang, 2015). In addition, these ten behaviors can be classified into two categories: learning behaviors and gaming behaviors. The coding scheme is shown in Table 1. To ensure the validity of the coding scheme, three experts who were experienced in game-based learning studies were asked to confirm the suitability of the codes and the corresponding definitions. In the second stage, the researchers set the trigger point of each behavior in the game based on the common mechanisms (e.g., battles, reading content, and escape from the battle) in the literature and practice. For instance, if the students go into the library in the game, click the book on the bookcase, and read the learning phrase from the book, the trigger point would be initiated, and would return the learning record of “R_P (reading the phrase)” to the database. To ensure the validity of the trigger points, one student who was not involved in this study played the game before the experiment. His learning behaviors were recorded automatically by the system, and the researchers performed a preliminary analysis of the records. Then, the experts mentioned above were invited to discuss and confirm the suitability of the trigger points and the coding result. The three experts agreed and confirmed the suitability and accuracy of the coding result based on the trigger points and the coding scheme. Finally, the trigger points were embedded in the game and automatically recorded and returned the students’ learning behaviors to the database.
### 實驗程序
1. 在實驗前，受試者有四個星期的英文基本訓練課程，是正規課程之一
2. 進行前測、學習動機問卷、英文焦慮問卷
3. 實驗開始：實驗組進行遊戲、控制組在老師用電腦與投影片上課，並且每人配發電腦，自行控制聽力的播放速度等等
4. 進行後測、學習動機問卷、英文焦慮問卷
5. 共有三個級別的學習內容，受試者在三週內完成這三個級別

In this study, the learning behaviors of students with different levels of English anxiety were explored. In order to examine the role of English anxiety in students' English game-based learning, a problem-based English listening game was developed for the students to learn while playing. Moreover, an experiment was conducted to ensure that the game was beneficial for the students' learning in some ways.
Before the experiment, the two groups of students had four weeks of classes about basic learning knowledge of English listening skills, which is a part of the existing English course. Fig. 5 shows the flowchart of the experiment. At the beginning of the learning activity, the students took the pre-test and completed the questionnaire of learning motivation and English anxiety. During the learning activity, the students in the experimental group learned with the problem-based English listening game. Their learning behaviors were recorded by the database of the learning approach for further analysis. On the other hand, those in the control group learned with the conventional technology-enhanced learning; that is, the teacher briefly presented the learning content and learning tasks to the students via a computer and a projector, and then provided the English listening practicing materials to them. During the practicing process, individual students were equipped with a computer to play the multimedia supplementary materials at their own pace. After the learning activity, all of the students took the post-test and completed the questionnaires of learning motivation and English anxiety.
From the experiment design, it can be seen that the learning materials for both groups were identical, while their learning approaches were different. The learning content was divided into three levels, and the students learned each level of the learning content each week. Finally, all of the students finished all of the learning content in three weeks.
## 結果
### 英文聽力成就分析
- 使用了ANCOVA
- 問題導向的英文聽力遊戲的確有助於學生的學習成效
In order to explore the students' learning behaviors in the problem-based English listening computer game, we first examined the effectiveness of the proposed approach to ensure that this learning approach could benefit the students' learning performance of English listening. Therefore, the one-way analysis of covariance (ANCOVA) was employed to evaluate the students' learning achievement in the experimental group and the control group. The assumption of homogeneity of regression was not violated (F ¼ 0.90, p > 0.05), indicating that the assumption was tenable and that ANCOVA could be used to interpret the relationships between the students' prior knowledge and their learning achievement in the post-test. Table 2 shows the results of the students' learning achievement in the post-tests. The adjusted means and standard error were 91.67 and 1.45 for the experimental group, and 83.03 and 1.47 for the control group. According to the statistics, it was found that the post-test score of the experimental group was significantly higher than that of the control group (F¼17.53,p < 0.001,h2¼0.19), which implies that the problem-based English listening computer game approach did in fact benefit the students’ learning of English listening. Furthermore, the effect size (h2) of learning approach was 1.92, representing a moderate effect size (Cohen, 1988).
### 學習動機的分析
- 使用了ANCOVA
- 問題導向的英文聽力遊戲提升學生的學習動機
In this study, there was a further attempt to ensure that the problem-based English listening computer game could improve the students' affect when learning English. Therefore, one-way ANCOVA was also used to compare the learning motivation of the students in the experimental group and the control group. The assumption of homogeneity of regression was not violated (F¼0.48,p>0.05), showing a common regression coefficient for one-way ANCOVA. Table 3 shows the results of learning motivation for the two groups. The adjusted means of the values were 4.40 in the experimental group and 3.77 in the control group. The standard errors of the two groups were 0.09 and 0.09, respectively. According to the ANCOVA result, it was found that the learning motivation performance of the experimental group was significantly higher than that of the control group (F¼26.24, p < 0.001,h2¼0.26). This result implies that the problem-based English listening computer game can more significantly improve students' learning motivation compared with the conventional technology-enhanced instruction.
### 英文焦慮分析
- 使用 ANCOVA
- 英文焦慮並沒有因為問題導向英文聽力遊戲而改變
This study also examined the effectiveness of the problem-based English listening computer game in terms of the students' English anxiety. Therefore, one-way ANCOVA was used to compare the English anxiety of the students in the two groups. The assumption of homogeneity of regression was not violated (F ¼ 1.00, p > 0.05), showing a common regression coefficient for one-way ANCOVA. However, the result of the one-way ANCOVA did not show a significant difference between the two groups (F¼1.67,p > 0.05). The adjusted means of the values were 2.99 for the experimental group and 3.16 for the control group. The standard errors of the two groups were 0.09 and 0.10, respectively. This result implies that the students' English-related anxiety cannot be easily changed by employing problem-based English listening computer games.
- 使用Ｔ檢定
- 英文焦慮無論在問題導向的英文聽力遊戲或一般傳統教學接不會因此而改善
Furthermore, the paired sample t-test was employed to examine the change in the students' English anxiety in individual groups. According to the results in Table 4, it was found that the students' English anxiety did not significantly change as a result of the treatment, regardless of whether they were in the experimental group or the control group. This result implies that students' English anxiety cannot be easily changed, either in conventional technology-enhanced instruction or in problem-based English listening computer games.
### 不同英文焦慮程度學生的學習行為分析
- 資料記錄
- 第一階段（聽然後選出圖片）的學生行為：9725個
- 第二階段（聽然後選出正確對話）的學生行為：7053個
- 第三階段（聽對話然後回答問題）的學生行為：7464個
- 低英文焦慮的人比較常做的行為
- G_D（在遊戲中探索）是三個階段中最常做的動作
- 第一階段中較關心 R_P（英語短語）
- 第二階段中較關心 L_S（完成學習任務）
- 第三階段中較關心 R_P（閱讀短語）
- 高英文焦慮的人比較常做的行為
- G_D（在遊戲中探索）是三個階段中最常做的動作
- 第一階段中較關心 R_T（閱讀聽力提示）
- 第二階段中較關心 L_S（完成學習任務）
- 第三階段中較關心 R_P（閱讀短語）
Several previous studies have reported the potential impacts of learning anxiety on students' English learning performance (Amiri&Ghonsooly, 2015; Na, 2007), while several researchers have indicated that game-based learning could help students face the challenges and fully engage them in the learning tasks (Palomo-Duarte et al., 2016; Richardson, 2016).
Therefore, in this study, we further examined the role of English anxiety in students' English listening practice, by analyzing the learning behavioral patterns of the students with high and low levels of English anxiety.
The students’ learning behaviors in the three gaming phases were recorded. A total of 24,424 behaviors were recorded, including 9725 codes in the first phase (i.e., the picture identification tasks), 7053 in the second phase (i.e., the listening and short answer questions), and 7464 in the third phase (e.g., the listening and comprehension tasks).
In addition, the students who learned with the problem-based English listening computer game were divided into those with lower and higher levels of English anxiety based on their responses regarding their English anxiety in the pre-questionnaire. The students with scores of English anxiety exceeding 3.00 in the pre-questionnaire were regarded as having higher levels of English anxiety (n¼18), while the others were categorized as having lower levels of English anxiety
(n¼21). Table 5 shows the frequency and percentage of the individual coded behaviors of the students with lower levels of English anxiety. It was found that “explore in the game context” (G_D) was the most frequent behavior in the three learning phases (52.46% in the first learning phase, 63.44% in the second, and 46.15% in the third). Meanwhile, the students with lower levels of English anxiety focused more on “reading the phrase” (R_P) in the first learning phase (14.01%), “finishing the learning tasks” (L_S) in the second learning phase (10.68%), and “reading the phrase” (R_P) in the third learning phase (29.27%). Table 6 shows the frequency and percentage of individual coded behaviors of the students with higher levels of English anxiety. Similarly, it was found that “exploring in the game context” (G_D) was the most frequent behavior in the three learning phases (53.77% in the first phase, 51.20% in the second, and 39.59% in the third). They also focused more on“ finishing the learning tasks correctly ”(L_S) in the second learning phase (16.07%), and “reading the phrase”(R_P) in the third phase (23.88%). However, in the first phase, they focused more on “reading the listening tips”(R_T) (16.54%).
- 圖片解說
- 每個編碼都是靜態，線上的數字代表 z-score ，有箭頭的線代表行為轉變，灰線是兩組皆有的行為轉變，而黑線是特定一組有行為轉變
- 第一階段行為分析
- 低英文焦慮者
- 傾向 R_T -> R_V -> G_S -> G_D -> L_S -> R_T （閱讀英文提示 -> 閱讀英文單字 -> 在戰鬥中獲勝 -> 發現遊戲中的下一個提示 -> 完成學習任務 -> 閱讀英文提示）
- 低焦慮者會在遊戲中不斷尋找英文提示，然後在戰鬥中獲勝，最後完成學習任務，之後重複不斷的取得下一關的成就。
- 他們比較享受遊戲
- 高英文焦慮者
- 傾向 R_V -> R_T，L_K -> G_E -> L_F （閱讀英文單字 -> 閱讀英文提示，取得秘密鑰匙 -> 逃離戰鬥 -> 學習任務失敗）
- 高焦慮者傾向先補充身邊的材料，在他們取得特定鑰匙時，會先逃離戰鬥，因此會沒有完成學習任務。由於他們的能力不足，時常會逃離戰鬥或戰鬥失敗
- 他們比較不享受遊戲
- 第二階段行為分析
- 低英文焦慮者
- 因為第一階段比較容易，因此學習行為較單純，但進入第二階段後，由於困難度提升，他們開始會出比較複雜的行為
- R_P -> L_F，G_E -> R_T，G_S -> L_K -> L_S -> R_V（閱讀短語 -> 學習任務失敗，從戰鬥中逃離 -> 閱讀英文提示，戰鬥成功 -> 取得秘密鑰匙 -> 完成學習任務 -> 閱讀單字）
- 高英文焦慮者
- 因為難度加深，因此又遇到了和第一階段一樣的問題
- L_K -> L_F（取得秘密鑰匙 -> 學習任務失敗）
- 第三階段行為分析
- 低英文焦慮者
- 出現一些反覆的行為
- R_T -> L_S -> R_T，G_E -> G_F，G_F -> G_E，G_F -> G_D，G_E -> G_S （閱讀英文提示 -> 學習任務成功 -> 閱讀英文提示，從戰鬥中逃離 -> 戰鬥失敗，戰鬥失敗 -> 從戰鬥中逃離，戰鬥失敗 -> 發現遊戲中的下一個提示，從戰鬥中逃離 -> 戰鬥成功）
- 高英文焦慮者
- 出現比以往還要更複雜的行為
- 當他們戰鬥成功後，會出現之前沒有的行為，包過再度戰鬥然後失敗、再度戰鬥然後逃離、閱讀英文單字、學習任務失敗等等
- G_S -> G_F，G_S -> R_V，G_S -> G_E，G_S -> L_F（戰鬥成功 -> 戰鬥失敗，戰鬥成功 -> 閱讀英文單字，戰鬥成功 -> 從戰鬥中逃離，戰鬥成功 -> 學習任務失敗）
- 另一個同樣的情況是，當他們取得秘密鑰匙後，可能會去戰鬥，然後成功或失敗，或者去發現遊戲中的下一個提示
- L_K -> G_S，L_K -> G_F，L_K -> G_D（取得秘密鑰匙 -> 戰鬥成功，取得秘密鑰匙 -> 戰鬥失敗，取得秘密鑰匙 -> 發現遊戲中的下一個提示）
- 另外當他們學習任務失敗時，會去取得秘密鑰匙，並且再次嘗試完成學習階段
- L_F -> L_K（學習任務失敗 -> 取得秘密鑰匙）
To probe the changes in the behavioral patterns of the students with different levels of English anxiety in the three weeks, a series of progressive sequential analyses was conducted to explore the changes. Fig. 6 shows the behavioral transition diagrams of the students with lower and higher levels of English anxiety over the three learning phases.
All of the sequences in the diagrams are statically significant. The values above each line represent the z-score for the sequence, while the direction of the line represents the direction of the behavioral transition. In addition, the gray lines represent the same behavioral transition of both groups of students in the same learning phase. On the other hand, the black lines represent the behavioral transition only shown in a specific group. For instance, in the first learning phase, both groups underwent the L_S/L_K transition, while only the students with lower levels of English anxiety underwent the L_S/R_T transition.
In this study, we first compared the behavioral patterns of the students with different levels of English anxiety in the firstlearning phase. It was found that the students with lower levels of English anxiety were inclined to read the listening tips and the vocabulary cards (R_T/R_V), use the vocabulary they had learned to win the battle (R_V/G_S), fight successfully and use their current knowledge to explore the game context (G_S/G_D), and complete the learning tasks and find another listening tip for the following tasks (L_S/R_T). On the other hand, the students with higher levels of English anxiety tended to read the vocabulary card and obtain the listening skills (R_V/R_T), get the key to finish the learning stage and escape from the following battle (L_K/G_E), and escape from the battle and then fail the learning tasks (G_E/L_F). This implies that the lower English anxiety students tended to solve learning tasks by seeking clues before reading the supplementary materials; in the meantime, they also enjoyed playing the gaming activities (e.g., battles in the game). In addition, they were willing to apply the acquired knowledge to more challenging tasks. On the other hand, the students with higher levels of English anxiety tended to read the supplementary materials before acquiring the listening skills. They did not enjoy the gaming tasks when completing the corresponding learning tasks. More importantly, their insufficient knowledge and gaming skills resulted in their failure in the learning tasks or battles.
In the second learning phase, the less anxious students began to undergo more complex learning and gaming processes,for instance reading the phrases and failing the learning tasks (R_P/L_F), escaping from the battle and checking the listening tips (G_E/R_T), fighting successfully and getting the key to finish the learning stage (G_S/L_K), finishing the learning stage and finishing their learning tasks (L_K/L_S), and finishing their learning tasks and consulting the vocabulary cards to understand the words from the following tasks (L_S/R_V). However, the more anxious students faced problems in the transition from finishing the learning stage to beginning the following learning tasks, as indicated by the correlations between finishing the learning stage and failing the following learning tasks (L_K/L_F). It can be seen that the lower anxiety students attempted to apply the learning knowledge they learned in the gaming context to solve their learning tasks and clear their gaming stage. When they failed, they returned to the learning stage and checked the learning tips. However, few similar behaviors were shown by the students with higher levels of English anxiety.
There were some interesting findings in the third learning phase. The students with lower levels of English anxiety had some sequential correlations including acquiring the listening tips and finishing their learning tasks (R_T/L_S, L_S/R_T), escaping from the battle and failing in the battle (G_E/G_F, G_F/G_E), failing in the battle and exploring more in the game context (G_F/G_D), and escaping from the battle and trying again to win the battle (G_E/G_S). On the other hand, those with higher levels of English anxiety performed more complex learning and gaming behaviors than they had previously.
Some behaviors involved when they fought successfully, including failing in the following battle (G_S/G_F), looking up the vocabulary cards from the following learning tasks (G_S/R_V), escaping from the following battle (G_S/G_E), or failing to finish the next learning tasks (G_S/L_F). Moreover, when they finished the learning stage (L_K), their behaviors of fighting successfully (G_S), failing in the battle (G_F), and exploring more in the game context (G_D) occurred. In addition, when they failed to finish the learning tasks, they tended to try again and finish the learning stage (L_F/L_K).
### 比較不同程度的英文焦慮者的學習成效
- 使用成對樣本Ｔ檢定
- 在高英文焦慮者身上，有顯著的進步，而在低焦慮者身上則有中等程度的進步
As reported by several previous studies, students' learning behavioral patterns could affect their learning achievements (Hsieh, Lin,&Hou, 2016; Tsai, Huang, Hou, Hsu,&Chiou, 2016); therefore, following the analysis of the effects of learning
anxiety on students' behavioral patterns, we investigated the effects of English anxiety on students' learning achievements. In this study, we first explored the change in the learning achievement of students with different levels of English anxiety. The paired sample t-test was employed to analyze the learning achievement of both groups of students (i.e., those with lower and higher levels of English anxiety), as shown in Table 7. For the less anxious students, the means and standard deviations were
88.33 and 11.63 for the pre-test, and 96.39 and 8.54 for the post-test. The paired sample t-test result indicated that there was a significant difference between the pre- and post-tests of the students with lower levels of anxiety (t¼"5.97,p < 0.001). On the other hand, the means and standard deviations were 67.14 and 20.28 for the pre-test, and 88.10 and 14.96 for the post-test for the more anxious students. The paired sample t-test result also indicated a significant difference between the pre- and post-tests of the students with higher levels of English anxiety (t¼"5.58,p < 0.001). The effect size (d) revealed a large difference between their pre- and post-test performance, while presenting a medium difference for the less anxious students.
- 使用 Welch's test
- 發現英文焦慮在前測對於學習成效有很大的影響，但在後側對於學習成效卻只有中等的影響
- 此外，雖然低焦慮者的表現皆比高焦慮者好，但在成績改善上，高焦慮者的改善幅度比低焦慮者還大
We also examined the effect of English anxiety on the students' pre-test and post-test results. Since the two groups of students (lower and higher English anxiety) are nested in this study, the Welch's test in the one-way heteroscedastic ANOVA was employed to control Type I error and power performance (Shieh&Jan 2015). The results of Welch's test indicate that the learning achievement (pre-test and post-test) of the two groups differed significantly, as shown in Table 8. The scores of the pre-test and post-test of the less anxious students were significantly higher than those of the more anxious students (F¼16.57,p < 0.001;F¼4.68,p < 0.05). According to the effect size analysis, it was found that there was a large effect size of English anxiety in the pre-test, but only a medium effect size in the post-test.
Fig. 7 shows the pre- and post-test scores of the students with lower and higher levels of English anxiety. It was found that the less anxious students performed better in both the pre- and the post-test. Meanwhile, it was found that the more anxious students improved their learning achievement more than the less anxious students did.

## 討論與結論
- 問題一、二，問題導向的英文聽力學習遊戲，比起傳統的應用科技技術的教學，有沒有對於學生的學習成效有所幫助？有沒有增加學生的學習動機？
- 遊戲式學習明顯對提升學生學習成效與學習動機優於傳統的教學模式
- 意味著使用英文聽力學習遊戲有助於學生學習，
- 問題四、五，高英文焦慮的學生和低英文焦慮的學生，在學習模式與學習成效上是否有不同之處
- 高英文焦慮者提高的學習成效要低英文焦慮者提高的幅度還大
- 意指高英文焦慮者在此系統下的受益較大
- 於遊戲過程中，高焦慮者比低焦慮者會逕行更多複雜的學習行為。例如：當他們適應了遊戲的運作與環境，他們會開始複習更多單字或獲取知識，用以解決他們在學習上或遊戲中遇到的困境
- 問題三，問題導向的英文聽力學習遊戲，比起傳統的應用科技技術的教學，有沒有減少學生的英文焦慮
- 研究顯示，使用此系統，學生的英文焦慮並沒有顯著降低
- 但此系統確實藉由複習學習教材與以前失敗的任務來有效練習英文聽力技巧，對於高英文焦慮者來說更是。這其中可能的原因為遊戲情景使學生處於輕鬆的環境，減少他們面對英文教材的焦慮
> 將學習內容嵌入遊戲任務中，學生學習上的焦慮能得到短暫的紓解，這對那些學習焦慮較高的學生而言是很重要的 [name= Harding (2000, pp. 77e86)]
> 遊戲策略使學生學習的更加快樂，且減輕學生學習上的焦慮，並鼓勵他們更加積極地獲取知識 [name=Beylefeld&Struwig, 2007; Shernoff,Csikszentmihalyi, Shneider,&Shernoff, 2003)]

- 而英文焦慮卻沒有降低的可能原因歸於幾類
> 學習時間不夠長 [name=Slavin, 2008]
> **這意味遊戲學習策略不夠有效率**

> 英文聽力比閱讀要有挑戰性 [name=Slavin, 2008]
> **學生英文焦慮的變化不夠明顯**

> 語言焦慮是一種跟學生語言能力相關的複雜且多維的感受 [name=MacIntyre&Gardner,1994; Peng, 2015]
> **因此探索英語焦慮的影響，需要考慮各種因素**

- 在本研究中，高焦慮者的學習行為與學習表現，在進步方面比低焦慮者要更好，因此將問題導向的英文聽力學習遊戲整合進電腦遊戲中做為教材，可以幫助學生克服對英文聽力學習的恐懼。
- 然而注意，本研究的一些限制
- 樣本規模過小，無法蓋推所有學生的情況
- 對照組的學生，在活動開始時，接受教師的指導，不是以自己的速度在學習，因此可能有淺在因素影響他們學習
- 在未來的研究中，最好為長期且大規模的實驗；可以考慮遊戲的內部整合；並且為了使學生的英文焦慮探究更加可靠，可以嘗試其他研究方法深入研究個人的思考方式，例如深入訪問個人感想等等。

Enhancing students' competence of applying foreign languages in reading and communicating has been emphasized (Khajavy et al., 2016). In Asia, English is regarded as an important competence, and some top-down promotion policies related to enhancing students' English abilities have been implemented (Huang&Chuang, 2016; Hwang et al., 2016). However, due to the limited learning time and environment, students have few opportunities to practice their English skills, and their English anxiety is difficult to eliminate (Hwang, Chen, Shadiev, Huang,&Chen, 2014). In order to better understand the role of English anxiety in students' learning, in this study, a problem-based English listening computer game was developed, and progressive behavioral analysis was employed to explore the differences in the learning behavior of students with different levels of
English anxiety.
According to the experimental results, the research questions of this study can be answered. Research questions 1 and 2 refer to the impacts of the problem-based English listening game on the students' learning achievement and motivation. It was shown that the learning achievement and learning motivation of the students learning with the gaming approach were significantly better than those of the students learning with the conventional technology-enhanced instruction. This implies that the proposed English listening game benefited the students' learning.
Furthermore, the analysis results of the learning behavioral patterns of the students with different levels of English anxiety provide further evidence to explain the findings.Research questions 4 and 5 refer to the differences between the learning behaviors and learning achievements of the lower and higher English anxiety students in the gaming activity. It was found that the students with higher levels of English anxiety improved their learning achievement more than did the students with lower levels of anxiety (as Fig. 7). In other words, the proposed approach benefited the more anxious students more than the less anxious students. Moreover, the progressive behavioral patterns also showed that, during the gaming process, the more anxious students attempted to engage in more complex learning and gaming behaviors than did the less anxious students. For instance, they started to review more vocabulary cards, and acquired knowledge to solve their learning and gaming tasks once they had adapted to the context of the game learning.
However, there was no significant difference between the English anxiety levels of the students learning with the gaming approach and conventional technology-enhanced instruction, which answers research question 3. On the other hand, the analysis results of the students' learning behavioral patterns further imply that although the students' English anxiety was not significantly reduced after the gaming activity, the gaming approach did engage them in trying their best to practice English listening skills by repeatedly referring to the learning material and solving the tasks they had failed before, which was particularly noticeable for those with higher levels of English anxiety. This could be due to the fact that the gaming scenarios situated students in enjoyable contexts, which then reduced their anxiety of facing the English learning materials. That is, by embedding the learning content in the gaming missions, the students' learning anxiety could be temporarily eased, which is especially important for those students with higher levels of learning anxiety, as indicated by Harding (2000, pp. 77e86).
Several previous studies have also reported that the gaming strategy enabled students to learn happily, and hence reduced their learning anxiety and encouraged them to more actively acquire knowledge (Beylefeld&Struwig, 2007; Shernoff,Csikszentmihalyi, Shneider,&Shernoff, 2003). This also explains the better learning achievements and more complex behavioral patterns of the more anxious students in this study.
On the other hand, the result of integration of the game-based learning environment did not decrease the students' English anxiety in the current study (research question 3). This result can be attributed to two reasons: one is that the learning time of the treatment was not long enough (Slavin, 2008), meaning that the effectiveness of the game-based learning strategy was insufficient. The other reason was that the learning goal of this study was to train the students' English listening skills as listening is regarded as more challenging than reading for students (Liu&Chu, 2010). Therefore, the change in the students' English anxiety was not obvious. Moreover, several researchers have claimed that language anxiety is a complex, multidi-mensional feeling associated with students' language (MacIntyre&Gardner,1994; Peng, 2015); therefore, in order to explore the impacts of English anxiety, various factors need to be considered. Although scholars have indicated that English anxiety may influence students' learning performance (Yen et al., 2015), in the present study, it was found that the students with higher levels of English anxiety showed more progressive learning behaviors and better learning achievement than those with lower levels. It is inferred that, integrating the problem-based English listening content into the computer game helped the more anxious students overcome their fear of listening to English, and hence improved their learning performance, as reported by Yan and Horwitz (2008).
It should be noted however that this study has some limitations. First, the sample size was small, and the results cannot be generalized to all students' perceptions in all circumstances. Second, at the beginning of the learning activity, the control group students learned the basic knowledge via the teachers' instruction; that is, in this portion of the learning process, they did not learn English at their own pace, which might have triggered some other factors which influenced their learning. In future studies, it would be better to conduct long-term and large-scale experiments with additional learning strategies following the present study. Furthermore, the intrinsic integration (Habgood&Ainsworth, 2011) of the gaming environment could be considered; in addition, to make the exploration of students' English anxiety more reliable, it is worth trying to explore individuals' perceptions using other research methods such as in-depth interviews or the think aloud method.