Keypal Friendships and Their Influences on Learner Development

Abstract

For teachers of English in countries where English is not used in everyday life, one major challenge has been creating opportunities for students to use English in meaningful ways for communication. In recent years, the development of computer-mediated communication (CMC) has provided ways of overcoming that challenge. For students like those in Japan where the authors teach, this can be a very important experience, since many students of English have rarely if ever had the opportunity to communicate in English outside of the classroom. In this paper, the authors will report on two pilot studies. The purpose of the studies was to examine aspects of keypal email exchange projects currently in development in English language classes at two Japanese universities: One purpose of these studies was to look at pedagogical aspects of the project, including implementation, and perceived and long-term benefits of such projects. Another was to look at interpersonal aspects, including the degree of development of the relationships formed and their influences on learner development and attitude. Results of this study indicate that about 3 out of 4 participants in each study planned to continue the relationship after the project ended and, although some of the students found the project difficult, most felt it had helped them with their English reading, writing, and vocabulary skills. Students also reported that the keypal friendship had influenced their views on foreign culture, their own communicative abilities, and comfort levels communicating with English speakers. This current research is but the first stage in a series of studies intended to examine Japanese university students' friendships formed with speakers of English via email and how these relationships affect their attitudes about their own English abilities and attitudes towards English in general. The researchers believe that such relationships can greatly enhance the learner's experience by personalizing and demonstrating the importance of English for intercultural communication.

Keypal friendships and their influences on learner development

In recent years, the development of computer-mediated communication (CMC) has allowed students of English as a foreign language, opportunities to experience communication in English, through email, chat, lists, and so on. In many cases, this is an opportunity that students have never had before, or have rarely had. It allows students to actually put to use the English they have learned and to improve their proficiency. It also opens up the possibility of forming friendships, which can further motivate students to learn English.

In this paper, the writers will discuss friendships formed as part of keypal email exchange projects and their influences on students' perceptions of culture and language skills. The writers will also report on an ongoing study examining these relationships along with the benefits and challenges of this type of project.

Literature Review

Email exchanges and Language Learning

Keypal projects can offer a number of positive benefits to the foreign language student such as exposure to new/authentic vocabulary and written expression; stimulation to explore new forms of language; ability to apply and practice latent knowledge; development of reading and writing skills; a decrease in foreign language anxiety as a result of personalized communication; development of friendships; and cultural awareness (Absalom &Pais Marden, 2004; Easton 2004; Greenfield 2003; Stockwell & Levy, 2001; Hopkins 1999; King & Orton 1999; Choi & Nesi 1998; Warschauer & Healey 1998; Mello 1998; Robb 1996b; Warschauer 1995). Moreover, the keypal project is ideal for encouraging language students to work autonomously and at their own pace: Students must initiate contact with individuals that they meet without teacher influence or aid; they must select language they feel is most appropriate; and they can recognize language deficiencies and attend to them to aid communication. When composing emails, students may look to a variety of resources for information such as texts, teachers, friends, the Internet or even the keypals themselves, thereby increasing the student's knowledge of the language and recognition of language resources.

One of the greatest advantages of the keypal project is the time students have to think about, create and respond to emails and as Lee (1999: paragraph 4) adds, "CMC makes it possible to provide individualized learning. Students can have time to read, understand and respond to the written message in text-based mode, thus having a chance to monitor and edit their or other participants' text-based dialogue structures." Students can choose topics they wish to discuss and practice, and rather than a teacher or student providing the response, the student receives real unrehearsed language to which they in turn must respond to. The language itself may be perceived as more accessible once communication is successfully initiated and cultural gaps are bridged.

Via CMC, students can identify paths to learning language on their own and hopefully will use these paths in the future, i.e. dictionaries, phrase books, translation devices, keypals, etc. Although direct vocal conversation does not play a part in CMC, Poster (1990) explains that it substitutes writing for spoken conversations and extends the domain of writing to cover areas of communication that previously were limited to face-to-face interactions, the mail, and the telephone, thereby enhancing reading, writing and vocabulary skills. Recently, however, with advances in technology, not only may students engage in synchronous typed chat with messenger services, but also synchronous spoken chat with services such as Skype and Paltalk. Students might begin their communication with emails and as their confidence builds, experiment with real time, spoken, or face-to-face communication.

Because students have the opportunity to learn and communicate without the direct aid of the instructor, a sense of personal accomplishment may be attained. Moreover, language practice is but one benefit of using modern technology to aid communication and understanding. Lee writes:


Marion Geddes, Gill Sturtridge, Rebecca L. Oxford, and Hana Raz (1990) also noted thatc[t]he process of learning how to learn pushes students to take responsibility for learning, to identify their preferred learning style. This process also helps learners to develop the skills of planning, arranging, and evaluating their own learning, to notice the specific strategies or behaviors (such as imagery, repetition, and naturalistic practice) they use in language learning to expand their strategy repertoire if necessary. (1999: paragraph 36)

CALL projects such as email exchange can offer students opportunities to examine how and what they learn so that the process of evaluating ability becomes a responsibility of the student. It introduces them to communication methods using modern technology and the target language.

Keypal projects with Japanese language learners

Tani-Fukuchi's 2005 study of Japanese learner psychology among university students suggests that these learners in particular have predominantly negative feelings towards their English study and prefer activities that are fun and practical rather than academically challenging, with a focus on communicative and practical learning opportunities. Additionally, more communication with native speakers was commonly listed as desirable by 1st year Japanese university students in a study by Carroll et al (2005) examining university curriculum design and students' reactions to it. In Brown's study of autonomous learning habits of Japanese EFL students, first-year college students recommended "talking or writing to foreigners" more than any other method of self-study to improve English skills (2005).

Takada's 2003 study of foreign language anxiety and McCroskey's 1985 study of communication apprehension experienced by Japanese students learning a foreign language, most common in the early stages of the learning process but also affecting more experienced learners, would suggest that a project such as email exchange, could help Japanese student language learners diminish such apprehension resulting in more confident and comfortable face-to-face communication. Such a project could also lead to positive feelings about the foreign language and culture in general, though a negative experience with email exchange could have the opposite effect.

Lee (1999: paragraph 23) writes "Under a text-based CMC setting, students can test their on-going learning process in a safer mode than in a face-to-face mode. They are free to take risks or a laid-back attitude, which is not recommendable, but sometimes necessary as a part of linguistic strategy, without feeling embarrassed not to 'talk'." It is this process of desensitization to foreign language communication which may be ideal for Japanese learners.

The Role of Interaction

While Krashen (1981) believed that comprehensible input was the necessary and sufficient condition for the acquisition of language, Ellis (1985: 161) concluded, based on an analysis of various studies and theoretical treatments of the subject, that both input and interaction influence second language acquisition. He lists eight characteristics of input and interaction which seem to facilitate rapid acquisition, based on this analysis. They are:

Of these eight, most are either facilitated by interaction or necessitate interaction, as opposed to input alone. Interaction can: 1) be used to elicit input, increasing its quantity, 2) help the learner control the propositional content, and 3) involve a range of speech acts, a high quantity of directives and extending utterances, and opportunities for practice. Making friends through interaction is one perceived need for communication and a realistic way to put language learning into context

Using Interaction to Learn Language

The opportunity to use language in interaction is an important part of language learning. Klein (1986:146-167) points out that the more the learner interacts, the more language he/she has an opportunity to learn, and the more language he/she learns, the more input he/she can solicit in order to learn more language. (Klein does not explicitly define interaction, but, based on the context, he seems to be referring mainly to spoken interaction.) At first, the learner uses whatever non-verbal means and small amount of verbal language he/she already knows. This elicits language, which the learner can use to confirm or disconfirm his/her hypotheses about the language.

According to Klein, learning language through communication is spontaneous (as opposed to guided) learning. The prototypical example of a spontaneous language learner is one who goes to another country to live and work or study. The learner carries out two separate but interconnected tasks -- to use his/her limited knowledge of the language to express him/herself and to understand others, and also to elicit input which helps in language learning. In spontaneous language learning, the focus is not on the language itself; it is on the content expressed by the language (Klein 1986: 15).

However, many language students do not have the opportunity to use English extensively outside of the classroom. They live in countries where English is not used in daily life, and where there are few English speakers to communicate with. This is referred to as a "foreign language" situation (as opposed to a "second language" situation, in which students live in a country where they use the target language in their daily lives).

In addition, language learners in a foreign language situation may have difficulty seeing English as a means of communication. In Japan, for example, English is generally necessary for college entrance exams, and students may think of English mainly as something that they learn about in order to take a test. The purpose of the language that they produce is mainly for the teacher to evaluate, not for the purpose of expressing their ideas. In such a setting, the Internet is important for language students, because it provides opportunities to use English for communication, to express one's own ideas and have interlocutors respond to the ideas themselves rather than to evaluate the language used to express the ideas. Davies, Shield, and Weininger (1998: 17-18) asserted:

There is a real possibility to interact "naturally" with native speakers... [communication] is real, despite the virtual interlocutors, whereas the target language interaction in the classroom is quite often unreal, despite the real interlocutors. Perhaps most importantly, though, is that the target language switches focus ...: language is no longer a goal but an instrument to pursue other (real) goals.

Even non-native English speakers who live in an English-speaking country may wish for more opportunity to use English through the Internet. In their daily life, for whatever reason, they may not be required to use English very much, or they may find communicating face-to-face in English stressful. Therefore, for English language learners in both EFL and ESL settings, the Internet can be a useful resource.

Dimensions of Friendship

In order to look at the relationships that Japanese students in an English language class formed as part of a project to correspond with native English speakers, we are using the dimensions of friendship identified by Parks and Floyd (1996). Parks and Floyd's dimensions measured development of relationships on six dimensions:

Purpose

In this pilot study, we have brought together interests in language teaching and interpersonal communication and the intersection between the two. For English-as-a-foreign-language students such as those in Japan, where there is little opportunity to communicate in English, this lack of opportunity can be a hindrance to language learning, both because communication in English can be an important source of motivation and because without it, students are deprived of the effects of interaction. Because the use of English to develop friendships is a motivating factor for some students (Kitao 1998), the writers are interested in looking at the relationships formed by English language students who correspond by email for a course project, and how this interaction influences learner development. Therefore, the purpose of this study is to examine: 1) the dimensions of friendships between Japanese college students and the people they correspond with for a course project, 2) students' perceptions of the project, 3) how they believe they benefit from it, and 4) how the development of the friendships influences learning skills and beliefs.

Methods

Overview

The following is a pilot study looking at the dimensions of friendship and their relationship to aspects of language learning. We surveyed two groups of students on their relationships and experiences with a keypal they were corresponding with as part of course projects. The questionnaire used in Study 1 included mainly short-answer questions about four of Parks and Floyd's (1996) dimensions of friendship, as well as demographic questions and questions about how long they had been corresponding with their partner. The questionnaire for Study 2 was adapted from the questionnaire for Study 1, but using a Likert scale to measure the dimensions, and with new questions related to demographics, students' English skills and skill development, motivation, opinion of the project, etc. This is intended to be an ongoing study so that future research will include case studies and a series of studies examining friendships and skills acquisition over time.

The Project

Instructions as to how to complete the project were given to students as part of their course syllabus and verbally reviewed in class. Students were asked to provide their instructor with 5 emails that they had sent to and 5 emails received from their keypals. Students were instructed to use an internet search engine to identify websites where they might initiate keypal friendships, but no other instructions, such as how to write the emails, were provided. Approximately 35% of the final semester grade was derived from the project.

Figure 1: The Assignment

Students reported using past instructors, international friends from religious, academic and social pursuits, pre-existing friendships and new friends found on the Internet to communicate with. Many students found it difficult to maintain relationships due to differences in schedules and keypal unresponsiveness. Due to these difficulties, students were encouraged to communicate with more than one keypal so as to ensure sufficient communication to complete the project.

This project was completely undertaken by students working autonomously. Students were not given topics to write about, nor did this writer edit or correct students' emails unless specifically requested by the student. Though non-responsiveness of the keypal was an issue early in the semester due to a range of difficulties, all but 3 of the students from Stage 1 (N = 71) and all of the students in Stage 2 (N = 19) completed the project. By giving students the freedom to choose their own keypals, and by allowing students to be completely responsible for the content of their emails, autonomy was emphasized. On several occasions, students brought their emails to class and asked their teacher to assist with a reply or translation. More often, however, students reported working with other students, using their own resources or asking the keypals themselves for guidance. Overall, students reported that though the project was extremely challenging, they really enjoyed the opportunity to use their English skills and foster new relationships.

Measurement Instruments and Procedures

The research was divided into two stages:

Stage 1

In order to gather information, 25 men and 46 women (N=71) were asked to complete a "Friendship Questionnaire" exploring students' feelings about their keypals and relationships with their keypals (Appendix A). This 12-item English-language instrument consisted of 1 Likert-scale and 11 free-response questions in English. Students were 1st and 2nd year English, Economics, and Business students at three universities in Kansai who had completed keypal projects over a 10-month period (an academic year). Students had a one-to two-week period to complete the questionnaires at their leisure. Students were instructed to write their answers in either English or Japanese, but were encouraged to use their English when possible. The teacher reviewed the questions with students to ensure understanding.

Stage 2

The questionnaire from Stage 1 was revised to both examine student's feelings about how their keypal experience affects their English skills and to gather additional information about the experience overall, as well as to measure the dimensions of friendship using a quantitative approach. This new instrument (Appendix B) included a new 25-statement 6-point Likert section (1= "I agree" and 6= "I disagree"), a 10-question open response section (asking most of the questions from the first instrument) and an expanded demographic section. Eighteen women and one man (N=19), average age of 19.5 yrs, from a second-year English reading class for English majors (Dept of Letters) were asked to complete this new questionnaire as a test group. Students only had one semester or 3 months to complete this project. Students had a one-to two-week period to complete the questionnaires at their leisure. Students were instructed to write their answers in either English or Japanese, but were encouraged to use their English when possible.

Results

Stage 1

The obtained results indicate a profile as follows: The keypal's average age was 23 years and nationality was North American (USA=47%, Canada=10%). The student and keypal communicated 78% of the time using primarily "email", for an average of 7 months. Students reported feeling "close" to their keypals (n=3.87, 1-7 point Likert scale) and 76% of the 68 respondents stated that they "will continue this friendship after the class is finished" (Figure 2.) for advice or support related to language or daily life (16%), and for friendship (46%).

Figure 2: Stage 1, Item 6 - Do you think you will continue this friendship after the class is finished? Why or why not?

Of the 70 respondents, 18% reported having met their keypals face-to-face at some time (students may have had relationships with their keypals prior to the start of the class project) and of those that hadn't met their keypal, 55% reported that they "maybe" or "would not" meet their keypals due to distance or fear, and 45% reported "wanting to meet" or "thought it was possible."

Of the 56 respondents, 64% reported that they didn't think they and their keypals "depended" on each other but for those 22 respondents who did, they reported depending on each other for matters of daily life (28%), language (23%), and advice and support (11.5%) (Figure 3).

Figure 3. Stage 1, Item 8 - Do you think you and your friend depend on each other? If so, in what ways?

Of the 46 respondents, 67% reported "discussing a wide range of topics" such as hobbies (12%), school (11.7%), culture and hometowns (7.3%) and family (6.8%). Additionally, 45% of the same respondents reported they "sometimes discuss personal topics" and 27% reported "being honest with each other".

Of the 41 respondents, 85% reported that they and their keypals did not "discuss [their] relationship or how [they] feel about each other", though 28% of responses included that students and their keypals were "good friends".

Of the 49 respondents, 61% reported that they didn't "know their friend well" and 29% of 39 respondents reported that students sometimes or usually have the ability to predict how the friend "will react to something or what their opinion will be" or "can usually tell what they are thinking."

Stage 2

Respondents reported communicating exclusively by "email" with 11 men and 7 women who were located mainly in the USA (73%). Students had emailed their keypal an average of 7 times over an average of 1.7 months. When asked to self-assess their "English skill level", students reported a mean of 3.4 on a 10-point Likert scale.

Results for the Likert scales.

Means for responses to the dimensions of friendship (Figure 4.) indicate that, although the relationships were not well developed on most dimensions (breadth was the most developed dimension), the participants saw potential in the relationships and hoped to continue them.

Figure 4. Means of the dimensions of friendship and commitment.

Means of students' perceptions of improvement in their skill, comfort, and confidence (Figure 5.) revealed that it was their vocabulary that students felt had improved most through this project. Not surprisingly, it was their speaking ability that they felt had improved least.

Figure 5. Means of students' perceptions of improvement in their skill, comfort, and confidence

Means of students' responses to the email project (Figure 6) indicate that in general, participants had a positive reaction to the project, even though they found it difficult. They did not feel that the project was a waste of time, and they liked exchanging emails and felt that it helped them improve their English.

Figure 6. Means of students' responses to the email exchange project.

Correlations among Likert scale responses. The correlations among the dimensions of friendship, the students' responses to the project, the students' perceptions of their improvement, the length of the relationship, the keypal's age, the number of emails sent, and the students' perception of their English skill level are reported in Figure 7.

Figure 7: Correlation Matrix

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

R

S

A

. Depth

.16

-.10

-.10

-.10

.43*

-.10

.38*

.45**

.49**

.43*

.24

.12

.13

.01

-.10

.21

-.20

-.10

B

. Dependence


.66

****

.72

****

.43*

.40*

.40*

.73

****

.19

.030

.24

.52

****

-.30

.59

****

-.40*

-.30

.30

-.30

-.40*

C.

Personalization



.70

****

.40*

.27

.37

.42*

.12

.21

.19

.54

***

-.40*

.49**

-.30

.44**

.26

-.40*

-.20

D.

Breadth




.56

****

.18

.62

****

.55***

.15

0.00

.20

.54

***

-.60

****

.71

****

-.40*

-.20

.33

-.20

.39*

E.

Commitment





.30

.45**

.16

.47**

.37

.55***

.54***

-.70

****

.83

****

-.50**

-.10

.18

-.40*

-.60

****

F.

Confidence






.49**

.64

****

.56

****

.60

****

.44**

.60

****

-.40*

.57

****

-.50**

-.20

.40*

0.00

-.60

****

G.

Vocabulary







.40*

.29

.16

.18

.58

****

-.70

****

.68

****

-.40*

-.20

.32

-.20

-.40*

H.

Speaking








.30

.17

.27

.60

****

-.30

.42*

-.40*

.02

.39*

.01

-.50**

I.

Writing









.80

****

.62

****

.73

****

-.40*

.40*

-.50**

-.10

.17

-.30

-.50**

J.

Reading










.50**

.53***

-.20

.36

-.40*

-.10

.13

-.10

-.30

K.

Comfort











.46**

-.60

****

.56

****

-.40*

-.40*

.56

****

-.60

****

-.40*

L. Help from Keypal











-.50**

.54***

-.40*

0.0

.16

-.20

-.60

****

M. Waste of Time












-.80

****

.67

****

.36

-.50**

.42*

.50**

N. St Likes Project













-.50**

-.30

.45***

-.40*

-.60

****

O.

Difficult Project















.35

-.30

.16

.54***

P.

Keypal`s Age
















-.40*

.44**

-.20

Q.

Length of Rel.

















-.30

-.20

R.

# of Emails Sent


















.28

S.

St Eng Skill Level



















St = Student, Eng = English

* = p<.05

** = p<.025

*** = p<.01

**** = p<.005

The correlations between the dimensions of the friendship and the measures of participants' perceptions of their improvement of skills, confidence, and comfort, and their feelings about the project were mixed. Depth of the relationship was correlated with all of the skills/confidence/comfort measures, except for perceived improvement in vocabulary. However, depth was not significantly correlated with any of the measures of positive or negative feelings about the project. The other dimensions of friendship did not seem consistently related to perceptions of improvement of skills/confidence/comfort. However, both commitment to the relationship and breadth of the relationship have significant positive correlations with positive feelings about the project (liking the project and feeling that it had helped their English proficiency) and negative correlations with negative feelings about the project (that it was difficult and a waste of time).

Except for depth, significant positive correlations were found among the dimensions of friendship. However, developing depth in these relationships may have been a problem, in part because of the short time involved but also because the students were aware that the teacher might read the emails. This might have restricted their willingness to write about personal topics, if they had been inclined to do so.

Significant positive correlations were found between development of confidence in using English and perceptions that individual skills had improved. There were also significant positive correlations with positive opinions of the project and negative correlations with the opinion that the project was difficult and a waste of time.

Participants who found the project difficult were less likely to feel that they had benefited from it, while students who liked the project and students who felt they had benefited were more likely to feel that they had improved their skills, confidence and comfort level. Students who liked the project and students who felt that they had been helped by it were also more likely to have well developed friendships along the dimensions we measured, except for depth (although since the relationships were of short duration, there was not much time to develop depth in the relationships).

A longer relationship did not necessarily lead to increases on the measures of the dimensions of the relationship, except for commitment (though since the relationships were all relatively short, it may have been too soon to measure this). The longer the relationship had gone on, the less likely students were to feel that it was a waste of time.

The age difference between the participants and their keypals ranged from five years older than the participant to six years younger. However, difference in ages did not seem to greatly affect the relationships, or participants' perceptions of the effectiveness of the project.

Since lower means on the Likert scale indicated that participants agreed, negative correlations indicated a relationship between the number of emails and other variables. Participants who sent more emails were more committed to the relationship and more likely to enjoy the project and less likely to feel that the project had been a waste of time.

Again, since lower scores indicated that participants agreed with the statement, while higher scores indicated a higher rating of their English proficiency, negative correlations indicated a relationship between the perception of their English proficiency and other variables. These correlations indicated that the higher participants' perception of their English proficiency was, the more likely they were to be committed to the relationship, the more they felt the relationship had made them more confident and comfortable, and the more they perceived it had improved their vocabulary, writing, and speaking. In addition, the higher participants' perception of their English proficiency, the less likely they were to feel that the project was a waste of time or that it was difficult, and the more likely they were to feel that they had been helped by the project and to have liked the project.

Responses to the open-response questions. All 19 students reported finding their friends on an Internet website. Japan Pen Friends, Yahoo, and e-pals were the only web sites that were mentioned specifically.

When asked whether they "will continue this friendship after the class is finished," 15 of the students (74%) reported that they wanted to. Of those participants, among the ones who gave reasons, 6 wanted to continue because they enjoyed doing it/it was fun/they liked to do it; 4 in order to improve their English, 3 because they liked the person they were corresponding with, and two for other reasons. (One participant gave more than one reason.)

None of the participants had met their keypal. Six of the participants wanted to, 1 wasn't sure, and 12 did not want to meet their keypal. For the most part, those who wanted to meet the keypals expected to have the opportunity; of those who said that they did not, 8 did not believe that they would have the opportunity due to distance, a lack of money, or a lack of time. Two of those who wanted to meet their keypal indicated that the keypal had specific plans to visit Japan in the future (in one case, the keypal plans to stay with the participant's family), and others mentioned that their keypal was interested in Japan and might someday visit. While some participants had no desire to meet their keypal, for the rest, whether they wanted to depended mainly on whether they believed it to be practical.

In answer to a question about whether they depended on their keypal, 10 of the participants indicated that they did not, and 2 answered "not yet." Several indicated that the reason that they did not depend on their keypal was that they did not know each other well enough (yet), or that they were not close enough (yet). Of the 6 who did answer that they depended on their keypal, most of the examples they gave were relatively superficial, such as giving advice related to finding jobs or Japanese and Japanese culture. However, since the relationships were relatively new, this was not surprising.

The topics participants listed as topics they discussed were: culture (7), family (7), life (4), future (4), hobbies (3), movies (3), music (3), job (2), literature (1), current events (1), pets (1), school (1), and weather (1).

Six of the participants said that they did not discuss personal topics with their keypals. The remainder discussed such topics as jobs (6), their future (3), love (1), family (4), school (2), hobbies (4), and where we live (1). Some of the topics did not seem personal, so some students may not have understood the question.

Six of the participants said that they had not talked about the relationship with their keypal. Some of the other responses were unclear as to whether they had discussed the relationship. However, 8 of the participants specified that they felt that they were "good friends" with their keypal, and 4 that they enjoyed the relationship.

Fifteen (79%) of the participants would recommend the project to other students who are studying languages, while 4 would not. (Interestingly, 3 of the 4 who would not recommend the project also did not feel that the relationship would continue after the class ended.) Among the reasons students gave that they would recommend the project were that it helped them improve their language skills (11), that it was fun or interesting (8), that they could form good friendships (4), and increased cultural understanding. As one student wrote, "If we get a friend in another world I become more curious about their country and language and try to understand them more."

The reasons they gave that they would not recommend the project were that they preferred face-to-face communication, that it was troublesome, and that the student did not have time to do it and did not feel that email was something she should be "forced" to do. Five of the participants, whether they recommended the project or not, said that they thought it was difficult.

In response to a question about what they had gained from the project, the participants listed improving language proficiency or confidence (10), learning about culture (6), forming friendships (5), and sparking a new interest in another country or culture (2). One wrote that she had learned that "There is differences between formal English on the textbook and usual conversations," and another wrote "If I wrong English grammer or spells, partner can understand almost. I have not afraided of wrong gradually. So, I think it is the most important thing to improve skills and make many friends!" Another wrote, "I think it is important to exchange cultures each other. Through this friendship I got to have interest in country of my friend and also my country. And I could learn that many foreigners have interest in Japanese and Japanese culture. I thought it is good to teach about my country to my friend." Even 3 of the students who did not recommend the project believed that they had gained something positive from it (learning about another country and culture using English positively).

Discussion

Stage 1 provided researchers with information regarding students and their keypal friendships. This information allowed researchers to then create a second instrument to better examine both interpersonal and pedagogical aspects of the keypal friendship. The 12 questions on the stage 1 questionnaire were complex and often attempted to examine more than one topic, making it difficult for student's to succinctly respond to all of the questions the researchers had. This was later rectified during the writing of the stage 2 questionnaire with the addition of Likert scale statements and the simplification of the questions from stage 1. Additionally, more demographic information was desired thus leading to stage 2 of the research.

The length of time spent by students exchanging emails with their keypals averaged 3 months in this stage of the research, and is probably not enough time for the students to have built a strong sense of rapport with their keypals, as evidenced by survey results regarding dimensions of friendship. However, the positive attitudes towards the project in general (willingness to continue the project, and general feelings about the project) indicate that the majority of students are able to recognize value in the project and the relationship with the keypal.

Though actual language skills were not measured in this research, students' responses to open-ended questions illustrate how the process of engaging in communicative tasks with native speakers of the target language can have positive affects on perception of skills, most likely resulting from the students' realization that they can in fact use their skills (at almost any level) to create meaningful exchanges with native speakers. It is this perception that directly influences how a student feels about the target language and his or her willingness to engage in tasks related to the target language. This could be described as an increase in confidence, though only slightly less than half of the students reported a positive change in their level of confidence in communication with foreign speakers.

Results of Stage 2 data illustrate that there is also a correlation between number of emails exchanged and student satisfaction with this project, though not necessarily between length of relationship and satisfaction. This would suggest that the amount of time actually invested in composing the emails directly reflects upon the satisfaction with both the project and the friendship. It is likely that as the student builds rapport with the keypal, he or she is more willing to spend time exploring possibilities for self-expression which lead to learning opportunities in the target language, illustrating Klein's theory of interaction.

Not surprisingly, the two strongest correlations occurred between perception of improvement in reading and writing skills, and commitment to the friendship and liking the project. We observed that student's emails were generally lengthier as the relationships progressed, and by their own report, became more interested in understanding subtle meanings conveyed in their keypals' emails as the rapport grew stronger. Students were often observed highlighting, underlining and circling sections of emails and a small group of dedicated students occasionally emailed or questioned their instructor for assistance with translation or composition. Because we are focusing more on attitude and perception, it remains to be seen whether skills actually did improve. Future stages of the research may include an objective measure of language skills pre- and post-project, but it would be difficult to illustrate that any changes on these measures were due solely to the relationship. Moreover, we would like to examine how the email exchanges over longer periods of time serve as a form of ongoing motivation to improve skills and are perceived as a confirmation of ability for the students. We are also interested in knowing how much time was spent by students composing their emails and what resources were used to get a clearer picture as to what the students are doing and how they are doing it.

As to the number of emails exchanged, though students were asked to complete 5 emails, they were also offered extra credit for their class grades if they completed additional emails. Because a semester is generally between 13-16 weeks, 5 emails has proven to be a manageable amount within the given time frame, though many students completed more than the required number. It is unclear if the motivation to complete additional emails was related to the grade for the class or for the enjoyment of communicating with the keypals.

More than half of the students in Stage 2 agreed that their keypal had helped them with their English, and we are interested in learning more about this cooperative function and how it is affecting the students' opinions of their abilities and motivation to improve their skills over a longer period of time. Students will be tracked via email and follow-up surveys will be administered in future stages of this research. Research questions we hope to answer include how the perceptions of skill improvement affect perception of future performance using language skills and how the keypal relationships affect ongoing development of learning skills and motivation.

There is a strong correlation between breadth of the friendship, reflected in responses to the statement "My friend and I discuss many different topics" and liking the project which would suggest that students who can write about a wide range of topics may be better able to enjoy the project, and do so sooner than students who have a difficult time expressing themselves. When students answer that their keypals help them with language, we would like to get a clearer picture as to how this happens, and how this "help" is affecting the process of rapport building and the dimensions of friendship. The current research does not address how the dimensions of friendship with the keypals differ from dimensions of friendship with regular friends. Likewise, the current research cannot reflect upon similar groups of students who do not complete keypal projects and their perceptions of skills. Therefore, student's perceptions about there regular friendships will have to be examined, and a control group of students who do not have keypals and their perceptions of skills will have to be compared to those students who do have keypals in future stages of research.

Although the majority of students were pleasantly surprised by the relationships they formed (as evidenced by short answer responses on both questionnaires), there were some who simply found the task stressful and "a waste of time". For teachers, it is important to identify those students early on and to spend time with them discussing how to make the project a successful one. The teacher may need to serve as counselor or coach to students who are greatly challenged by the project. Future stages of this research will examine in greater detail the difficulty students encounter and how the teacher can assist in these circumstances. Moreover, this difficulty highlights the fact that this project is unique in that the students find themselves dealing with both real individuals and issues involving intercultural expectations and communication. Problems for students, such as non-responsiveness of the keypal, not having much in common, or simply not enjoying their experiences with a specific keypal (just as students do not form a friendship with everyone that they meet face-to-face), warrants the need to explain to students that they may need to initiate more than one friendship so that they may meet the requirements of the project and do so in an enjoyable manner. In these cases, we would like to explore how negative social experiences affect learner attitudes towards the target language and culture.

Though not assessed by the questionnaires discussed within this paper, it is important to reiterate that instructions for the project be clear and concise. Although a certain level of autonomy is advisable in terms of how students identify keypals and who these people are, we found that providing students with clear examples (as is currently in practice) of English paragraphs, grammar, sample questions and responses, and examples of websites that may be used to find keypals, is also advisable. Additionally, requesting a minimum number of sentences per each student email is advisable to ensure that students are practicing as much English as possible and not just writing one sentence per email. Students should also be encouraged to practice English learned in the classroom with their keypals as an opportunity for practical application of skills and opportunities for authentic responses.

A number of students commented about the lack of privacy involved in their relationship with their keypal because they were required to submit all emails to their teacher to ensure they had adequately completed the project. One student wrote: "the content of the emails was tailored for handing in as an [sic] project." Therefore, it is important to discuss with students what kind of privacy policy is most effective and desirable. This may depend on the age and culture of the students. As students who completed questionnaires for the current stage of this research had to hand in their emails to the teacher, the results regarding the dimensions of friendship may have been affected by the student concerns for privacy.

Ideas for Future Research

Stage 2 of the research served as a pilot study to examine the new instrument created. The new questionnaire allowed for easier analysis of results and more discrete questioning of specific dimensions. It also allowed for statistical analysis of the results. This instrument was recently converted to an excel file and emailed to 180 students at two universities - Stage 3 of the research. The writers wanted the entire process of the project, both the emails and the questionnaires, to be completed using the computer and Internet to enhance the CALL process. Additionally, using email to distribute and collect questionnaires made the capture and transfer of data into excel files easier for the researchers. Students have been asked to return their completed questionnaires via email and that data is currently being analyzed for the next stage of this research.

Researchers are considering using a web-based survey method to more easily capture information. Additionally, we are preparing to create additional surveys to examine friendships between students and their regular (vs. keypal) friends, control groups (students at similar levels and ages) of students who do not complete keypal projects so as to examine their perceptions of skills, and students who had strongly negative and positive experiences with the project so as to get a better sense of what students are experiencing cognitively and emotionally during the email exchange process. It is our hope that these additional evaluations will help shed light on many of the questions raised by our initial results.

Conclusion

Due to recent innovations in technology, instructors of foreign languages have opportunities to encourage autonomous learning and application of skills like never before. Robb (1996a) gives some excellent advice to prospective project organizers:

Key pals can be an extremely rewarding experience for your students, but don't expect everything to go well the first time. Just like any other aspect of your teaching, it will take some experience to discover the best implementation for your own curriculum. Even with first-time glitches, however, you can be sure that it will be an experience that your students will not forget. Don't be surprised to find some students exchanging snail-mail addresses with their Key pals, turning a virtual friendship into an actual one. It happens! (paragraph 20)

Robb`s statement about friendships is precisely what the writers hope will happen and what a few of our students have already done. Two of the students have reported meeting their keypals either by travelling abroad, or by hosting the keypal in the student`s home in Japan. Both of these students reported strong changes in perceptions of their skills and added that these changes were directly related to these experiences.

As a result of the current research, we hope to further examine constructs such as motivation and confidence related to self-perceptions of skills and how they are influenced by the keypal friendship. Ongoing tracking and evaluation over the students' university career will offer insight into long-term effects of the project and case studies with individual students will offer a richer picture as to reasons why these types of projects succeed or not for individual students. The addition and comparisons of control groups of students who do not complete these projects will certainly offer some information regarding the differences in attitudes towards learning and language of students' who enter into friendships with foreign speakers.

Though our time as teachers may be limited to a semester or two with our students, initiating a project such as this may offer benefits to our students that we could never provide in a classroom setting. Giving our students the ability and motivation to use the language skills they have already acquired, in real-life situations, is an opportunity that can help diminish communication anxiety, educate students about culture and initiate friendships that can rekindle motivation to learn and last a lifetime.

References

Absalom, M., Pais Marden, M. (2004). Email communication and language learning at university - An Australian case study. Computer assisted language learning, 17(3-4).

Brown, R. (2005) Autonomous learning strategies of Japanese EFL students. The language teacher, 29(4).

Carroll, M., Douglas, L., Harrison, M., &Tsurii, C. (2005). Curriculum: What teachers do and what do students think of it? In K. Bradford-Watts, C. Ikeguchi, & M. Swanson (eds.) JALT 2004 conference proceedings. Tokyo: JALT.

Choi, J. and Nesi, H. (1998). An Account of a Pilot Key Pal Project for Korean Children. English Language Teacher Education and Development. The Centre for English Language Studies, The University of Birmingham and The Centre for English Language Teacher Education, The University of Warwick. 4 (1), Autumn. Retrieved July 27, 2004 from http://iteslj.org/Articles/Choi-KeyPals/index.html

Davies, L.B., Shield, L., &Weininger, M.J. (1998). Godzilla can MOO, can you? MOOs for construction, collaboration and community and research. The language teacher, 22(2), 16-21.

Easton, E. (2004). Retrieved July 27, 2004 from http://eleaston.com

Ellis, R. (1985). Understanding second language acquisition. Oxford: Oxford University Press.

Greenfield, R. (2003). Collaborative email exchange for teaching secondary ESL: A case study in Hong Kong. Language learning and technology. 7 (1), 46-70.

Hopkins, G. (1999). Getting Started on the Internet: The Key to Keypal Success! Retrieved July 27, 2004 from http://www.education-world.com/a_curr/curr125.shtml

King, L. and Orton, C. (1999). Action research project: A study of motivation in keypalling in the CEGEP ESL classroom. Retrieved July 27, 2004 from http://www.magma.ca/~leking/keypalling.htm, paragraphs 1, 49-50.

Kitao, S. K. (1998). Interaction and On-Line Synchronous Communication in English Language Learning. CALL-EJ online, 3(1). Retrieved August 20, 2004 from http://www.clec.ritsumei.ac.jp/english/callejonline/3-1/kkitao.html

Klein, W. (1986). Second language acquisition. Cambridge:Cambridge University Press.

Krashen, S.D. (1981). Second language acquisition and second language learning. Oxford: Pergamon.

Lee, C. (1999). An exploration of pedagogical implications of using networked-based computer mediated communication in the communicative language classroom. Interpersonal computing and technology: An electronic journal for the 21st century. October 1999 - Volume 7, Number 1-2. Retrieved July 27, 2004 from http://www.emoderators.com/ipct-j/1999/n1-2/contents.html

LeLoup, J.W. and Ponterio, R. (1995). Addressing the Need for Electronic Communication in Foreign Language Teaching. In Richard Steinfeldt (ed.), Educational technologies, monograph of the New York State Council of Educational Associations. 39-54.

McCroskey, J.M., Gudykunst, W.B., &Nishida, T. (1985). Communication apprehension among Japanese students in native and second language. Communication research reports, 2: 11-15.

Mello, V. (1998). Report on a Keypal Project, and Tips for Keypal-Project Success. The Internet TESL journal, 4 (1). Retrieved July 27, 2004 from http://iteslj.org/Techniques/Mello-Keypal.html

Parks, M. &Floyd, K. (1996). Making friends in cyberspace. Journal of computer-mediated communication, 4(1). Retrieved October 31, 2004 from http://www.ascusc.org/jcmc/vol1/issue4/parks.html

Poster, M. (1990). The mode of information: Poststructuralism and social context. Chicago: The University of Chicago Press.

Robb, T. (1996a). Email keypals for language fluency. Retrieved July 28, 2004 from http://www.kyoto-su.ac.jp/~trobb/keypals.html

Robb, T. (1996b). Foreign Language Notes (Foreign Language Educators of New Jersey), 38 (3), 8-10.

Stockwell, G., Levy, M. (2001). Sustainability of email interactions between native speakers and nonnative speakers. Computer assisted language learning, 14(5).

Takada , T. (2003). Learner characteristics of early starters and late starters of English language learning: Anxiety, motivation, and aptitude. JALT journal, 25(1), 5-29.

Tani-Fukuchi, N. (2005). Japanese learner psychology and assessment of affect in foreign language study. The language teacher, April, 2005, 29(4), 3-9.

Warschauer, M. (1995). E-mail for English teaching. Alexandria, VA: TESOL Publications.

Warschauer, M. and Healey, D. (1998). Computers and language learning: an overview. Language teaching, 31, 57-71.

Appendix A

Stage 1 Questionnaire

Appendix B

Stage 2 Questionnaire

Biography

Marlen E. Harrison is a lecturer at Momoya Gakuin and Doshisha Universities in Western Japan. He received his MA in Education and Human Development from The George Washington University. His current research interests include testing and assessment, keypal projects, and social constructivism in the language classroom. He is the current Co-Coordinator of JALT's Learner Development Special Interest Group.

S. Kathleen Kitao is a professor at Doshisha Women's College, Kyoto, Japan. She received her MA and PhD in communication from Michigan State University. She is a co-author of more than 50 textbooks. They include composition, reading, speaking and vocabulary development textbooks. She is also the author of several books, including Theory and Application in English Language Teaching and Reading, Schema Theory, and English Language Learners. Her interests include linguistic pragmatics, interpersonal and intercultural communication, issues of research methodology, and language teaching, and specifically using computers in language teaching.