Online intercultural collaborations using wikis : An analysis of students ’ comments and factors affecting project success Edo Forsythe Hirosaki

Hirosaki Gakuin University forsythe@hirogaku-u.ac.jp Online intercultural collaborations using wiki pages are becoming common in foreign language education. However, research into their effectiveness and the types of comments students make in their interactions remains scant. This study addresses this gap by exploring three years’ worth of interactions among Japanese students of English and American students of Japanese. The data demonstrated that the focus of students’ comments matches the overall topic of the collaboration, that interpersonal topics generate more interactions, and that logistical factors have a strong impact on the success of online collaborations. Based on these data-driven conclusions, this paper provides lessons learned and specific tips for conducting successful, online intercultural exchanges among foreign language students using wiki pages as a collaboration medium.


introduction
Online interactive collaborations have become much more prevalent in foreign language classrooms in the past decade as technology and bandwidth have enabled instant access to people and places around the world.Foreign language classrooms have long struggled with the problem of isolation from the target language and culture -this has been especially true for Asian students of English and English-speaking students of Asian languages.Thanks to the Internet and Web 2.0 tools, this problem exists no more.
One might ask why interactive collaborations have become so popular in foreign language (l2) classes; there are several answers.First, conducting computer-based intercultural exchange activities provides students with an opportunity to develop the communications and computing skills necessary for work in the 21st century global market.The Independent Data Corporation (2013) and Schenker (2013) both found that the top skills required for the best jobs in the 21st century workplace are oral and written communication skills as well as computer literacy.Online intercultural collaborations offer a chance for students to hone both of these skills: the communications opportunities provided include a native speaker with whom language learners can practice their l2 skills as well as learn about the target language culture from a first-hand source.Internet-based intercultural exchanges are especially valuable for language learners living far from the country in which their l2 is spoken, such as Japanese learners of English or American students of Japanese (Nielsen, 2013).
Possibly the most impactful rationale for engaging students in online intercultural collaborations is the fact that research has shown that students feel less pressure when interacting online than in face-to-face situations, and therefore are more active in their interactions and take more turns in communicating with others (Brandl, 2012;Ito, 2011;Koh & Lim, 2012;and Nielsen, 2013).Anything that teachers can do to lower their students' affective filters and improve their engagement must be acted upon.People interact online with more anonymity than in face-to-face situations, and therefore, feel less afraid of making mistakes in their communications.Providing language learners with a safe environment in which to practice their new language without fear of ridicule is a priority for foreign language teachers.Online interactive collaborations provide precisely the environment needed.
This paper reports on the findings of data collected through an online intercultural collaboration project conducted over a three year period between Japanese students of English at Hirosaki Gakuin University and American students of Japanese at the University of Wisconsin -River Falls campus.All of the project wiki pages and comments can be viewed at <http://aomoriuwrf.pbworks.com>.An explanation of the project's beginning and conduct from 2011 to 2013 will be provided along with a discussion of the success of the communications based on the presence of unanswered questions (as in Tanaka-Ellis, 2011), and then a review of the types of comments and interactions conducted by the students in both their native language (l1) and their l2.Finally, lessons learned will be shared with suggestions of how language teachers can begin successful, online intercultural collaborations in their own classrooms.

Platforms
Since technology has become a normalized part of the language education process (Bax, 2012), Internet-based collaborations have used a variety of platforms: blogs, podcasts, wiki pages, and social networks.The pedagogy behind these collaborations is similar: to provide learners with an opportunity to use their l2 in a realistic setting with native speakers while performing a task or learning about the target language culture.Among available collaboration platforms, wiki pages have been found to be the best-suited for collaboration (Chu, Siu, Liang, Capio, & Wu, 2013;Koh, & Lim, 2012;Miyazoe, & Anderson, 2010;and Nielsen, 2013).Miyazoe and Anderson (2010) investigated student perceptions in online interactions and found that students prefer collaborations through wiki pages versus blogs or forums.This preference for wikis over other platforms is most likely due to their simple, flexible nature and their ease of creation.Students can begin creating their own wiki pages with minimal instruction and wikis' ability to accommodate a variety of media allows students several options in their collaborations.Wiki pages allow students to use pictures, videos, audio files, hyperlinks to external websites, or simple graphics to share information needed to perform their assigned task, and are therefore, very well-suited for online intercultural collaborations.(See Chu et al., 2013, for a detailed explanation of the uses of wikis in higher educational contexts for knowledge management.)

Collaboration group dynamics
As online interactive collaborations have become more common in the past five years, educators have begun analyzing various aspects of their conduct and the learning outcomes of the participants.Ito (2011), Koh and Lim (2012), Lee (2011), Oliveira, Tinoca, and Pereira (2011), and Wang (2011) all explored the aspects of group work and group dynamics in online collaborations.With regard to the size of groups which best encourage robust online interactions, Oliveira et al. (2011) and Wang (2011) all found small groups preferable to large groups.Small groups encourage the participation of all group members and makes it more difficult for some group members not to participate.Additionally, teacher participation in the group activities was found to have little impact on the types of group interactions (Nielsen, 2013); however, the presence of an instructor in the collaboration can spur interactions among group members who are having difficulty in beginning their collaboration (Oliveira et al., 2011).Therefore, it behooves teachers to monitor the groups' online collaborations to ensure that progress is being made and to provide encouragement when groups are having difficulty beginning their task.

Interaction types
When planning online collaborations between students in different countries or regions, teachers have two interaction options to choose from -synchronous or asynchronous.Synchronous interactions can be quite logistically challenging to set up due to the technology needs and the difficulty of aligning time schedules across time zones.Also, synchronous exchanges can have some negative effects on the students, such as increased pressure to perform, higher stress levels, and an inability to expand their l2 knowledge because of a need to respond immediately to interactions using only their existing language skills (Brandl, 2012;Ito, 2011;and Jung, Kudo, & Choi, 2012).
In contrast, asynchronous collaborations have been demonstrated to be overwhelmingly beneficial to all students, especially beginners (Chun, 2011;Hirotani, & Lyddon, 2013;Ito, 2011;Nielsen, 2013;and Tanaka-Ellis, 2011).Asynchronous collaborations allow students to plan, create, and revise their work prior to posting it to the collaboration platform; often the posts are much longer than those seen in synchronous interactions (Chun, 2011).Having time to thoroughly consider their input affords students the opportunity to expand their language skills because they can learn new vocabulary or new grammatical constructs which allow them to say what they want to in the interaction (Hirotani & Lyddon, 2013).Having time to consider and plan one's posts is especially beneficial for beginning language learners who may not have developed sufficient language abilities to express themselves as required by the collaboration tasks (Nielsen, 2013;Tanaka-Ellis, 2011).

Task types
There are a variety of tasks which can be assigned in intercultural online collaborations.Common examples are interpersonal exchanges, sharing cultural information, and collaborative project work in teams or small groups.When deciding the type of tasks to assign, it is important for the instructor to keep in mind that they are creating a learning community for their students to learn both on their own and with the help of others (Wang, 2011).Jung et al. (2012) found that instructions for task completion should be stated as clearly as possible to lower the students' affective filters and stress levels, while increasing the potential for successful task completion.Brandl (2012) found that open-ended tasks encourage more interaction and participation among the students.Simpler question and answer-style tasks are limiting in their potential for students to expound upon the topic at hand; so instead of providing a set of questions to be answered through interviewing a partner about an aspect of their culture, the students should be tasked with researching and reporting back on a more general topic.For example, Japanese students of English could be assigned to compare the traditional celebrations of the New Year holiday in America and Japan.Such an assignment forces students to contemplate their own culture's practices and engages their critical thinking skills to determine what questions to ask in order to elicit data for a comparison of the topic.
In analyzing how students responded to their tasks and the language they used, Brandl (2012), Lee (2011), andTanaka-Ellis (2011) found that students rarely corrected one another's linguistic errors within the collaborations, nor did students' language improve when their errors were pointed out by others.These findings suggest that online collaborations are more beneficial for learning about target cultures than for directly improving one's language abilities; however, the positive effects of the opportunity for students to interact with native speakers cannot be discounted.Online intercultural collaborations provide students with the chance to interact with native speakers in their l2 and thereby developing greater confidence in their l2 interactions -an opportunity which they may not have within their own communities.This alone is a powerful rationale for conducting such online collaborations.
In summary, a review of the growing number of articles investigating the pedagogical impact of online intercultural collaborations shows that wikis are a preferred platform for collaboration among both teachers and students due to their flexibility and ease of use.When planning the collaboration, teachers should keep in mind that small groups allow for more engagement and interaction among the members; and asynchronous collaborations provide an excellent opportunity for students to plan their interactions and improve their language skills while lowering the stress they experience from interacting in a foreign language.Finally, open-ended tasks allow for freer discussions and the ability to express off-topic thoughts while collaborating with their teammates on a project.Even though students focus primarily on the content of the task assigned, the fact that they are engaging with native speakers in their l2 builds their l2 confidence while they learn about their target country's culture and people.
A review of the literature revealed no studies which specifically explore the types of comments made by participants of online intercultural exchanges.Language use, pragmatics, and communicative competence have been investigated, but little research has been conducted on the content of the students' interactions.The current collaboration project and analysis of the data were conducted with these ideas in mind.

Participants and procedure
The students of Hirosaki Gakuin University who participated in this study's intercultural collaboration were freshmen students in the university's School of Liberal Arts' Department of English and English Literature.All of the participants were English majors.In all three years, the Japanese students began the collaboration after the American students due to an uneven alignment of the school calendars.The collaboration was an extra activity for the Japanese students, which was assigned as homework due to the unavailability of a computer lab during the class meeting time.Not all of the students followed the assignment schedule, but the majority of them did.
The American participants were students from the University of Wisconsin-River Falls (uwrf) in a third semester Japanese language class (japn201).Technology is highly integrated in the daily instruction as well as the course assignments and projects.japn201 students have done a similar wiki-based collaboration project among other uw campuses and with other Japanese language students during the second semester Japanese course.The wiki-based collaboration with the Japanese university students was a part of the course project and graded.For this collaboration project, the American students were required to create a wiki page on a given topic in Japanese and read the Hirosaki Gakuin students' English wiki pages and leave a comment in English.During the first year of the collaboration, the Japanese students were assigned to create a pictorial culture project in English.They were asked to explain an aspect of local culture along the lines of the Flat Stanley project (https://www.flatstanley.com)using the Hirosaki city mascot, Takamarukun, in place of Flat Stanley (see Figure 1).Alternatively, the Uwrf students adopted a group collaboration project approach.Students were divided into small groups and each group was assigned a different theme (see Table 6 below) and created a group wiki page in Japanese.During the second and third years of the intercultural collaboration project, both the American and Japanese students were assigned to create wiki pages about themselves to introduce themselves to their counterparts (see Figure 2 for an example of the student's wiki pages).In all three years, the students were instructed to read their fellow group members' pages and leave comments in the page author's l2 about the content -the focus was on sharing cultural information instead of correcting one another's l2 language errors.
As a means of integrating the collaboration project into the curriculum and to give the Japanese students a reason for participating, in all three years of the collaboration project, the Japanese students' final exam (an oral proficiency interview) included a discussion of something interesting about American culture that they learned from the collaboration wiki pages.This did not require them to create their own page, but did cause the majority of the students to at least visit and read the pages -although there were some students who admitted during their final examinations that they did not look at the wiki pages prior to the examination.

Grouping of participants
Because it might have been daunting for the participants of this exchange to attempt to comment on all of the other participants' wiki pages, it was decided to create small groups of participants to facilitate more interactive communications.Due to uneven student numbers, the participants were randomly assigned to groups usually consisting of two or three Japanese students and one American student.The participants were allowed to view and comment on any other wiki page in the collaboration, but they were instructed to first view and comment on the pages of their group partners.Using this structure provided the participants with a structure within which they could begin communicating, and allowed participants to expand their attention to other wiki pages once they gained confidence.
Organizing the participants into small groups was found to be in line with the research findings and recommendations of Oliveira et al. (2011) and Wang (2011).

Research questions
In keeping with Tanaka-Ellis' (2011) analysis of participants' comments in intercultural collaborations, the following research questions were proposed in this study's review and analysis of the students' wiki pages and their comments: a. Do language learners attend to previously posted comments when participating in an online intercultural collaboration? b.What are the possible causes for questions posed in wiki comments being left unanswered?c.What types of intercultural communications can be expected during online collaborations among language learners in different countries?d.Does the theme of the intercultural collaboration have an impact on the types of comments posted by participants?

Data analysis
The data from the participants' comments on the collaboration wiki pages were analyzed in two ways.First, a quantitative review of the numbers of comments was conducted.This data was then analyzed from different perspectives and with comparisons of related comment types: the numbers of comments by language, the numbers of author responses to others' comments, and the numbers of comments posted by the two instructors.Second, a qualitative analysis of comment types was performed to determine the types of comments being made by participants in online intercultural collaborations.The comments were read and divided into three categories: interpersonal, cultural, and language-related.As no other research was found which analyzed the types of comments that language students make in intercultural communications, the author chose to categorize them based on the general topics observed.In situations where a single comment posted contained both interpersonal information and cultural statements, the comment was counted in both categories.In future analysis of participants' comments, it may be more prudent to analyze the comments at the sentence level instead of viewing them as single comment posts.

Analysis of unanswered questions
In reviewing the numbers of questions asked and answered or unanswered as an indication of actual communications and intercultural exchange, the data (see Table 2) showed that a total of 59 questions asked on the Japanese-language wiki pages and 115 questions were asked on the English-language wiki pages over the three-year term of the collaboration.Attending to previous posts in learners' answers.Overall, the comments on the Englishlanguage wiki pages included twice as many questions as the Japanese-language wiki pages; however, the percentage of unanswered questions did not have the same proportions.In reviewing the question patterns in individual years, the data in the first year show a significantly different pattern than in the succeeding years.On the English pages, over half (n = 62) of the entire quantity of questions posed occurred in 2011.Also, the highest percentage of unanswered English-language questions (58%) was posed during the first year of the collaboration.Conversely, the quantity of questions posed on the Japanese pages in 2011 was only 5% (n = 3) of the total number posed in the three-year term of the collaboration and only one of them went unanswered.
There are several possible reasons for these results.In the case of the collaboration in 2011, timing seemed to be the primary factor in the lack of communications indicated by the low number of questions on the Japanese-language pages and the high number of questions on the English-language pages.In the initial collaboration, the Japanese participants did not create their wiki pages until late in the semester: the teacher's example page -Aomori Project -was created on November 17, 2011 and the student-created wiki pages were all posted after December 4, 2011.In the same year, the American students had all created their pages in Japanese by November 15, 2011 and were waiting anxiously for the Japanese students to create their pages in English so that the collaboration could begin.By the time that the English-language pages were posted, only two weeks remained in the American participants' semester so there was little time for back-and-forth communication via the wiki pages and the American participants' great number of comments and questions may have overwhelmed the Japanese participants and forced them to focus more on responding to comments on their own pages instead of asking questions about the content on the Japanese-language pages.This could also have been due to the Japanese participants' lack of confidence in using technology and a general lack of experience in using the Internet or computers in their educational process (see Murray & Blyth, 2011;Nielsen, 2013, for additional information about Japanese students' lack of technical savvy).
This imbalance in question posing and responding in the 2011 collaboration was mirrored in the total number of comments posted on the wiki pages (see Table 3).Again the English-language wiki pages exhibited more interaction than the Japanese-language pages for possibly the same logistical reasons mentioned above.The 2011 English-language pages had more comments and responses than in the other two years combined in spite of the shortened time limit imposed by the university calendar.In addition, while there were more comments overall on the Japanese-language pages in 2011, the number of author responses was less than half of the amount seen in the 2012 and 2013 collaborations.This may indicate that since the American and Japanese participants were focusing on the English-language pages, the Japanese participants had little opportunity to post questions to the Japanese-language pages, and therefore, the American participants had less motivation to respond to comments on their own pages.The higher percentage of author responses on the English-language wiki pages would seem to indicate that the focus of the collaborative communications remained on the English-language pages on which the American students pounced as soon as they were created, thereby not really allowing the Japanese students a chance to fully delve into the content presented on the Japanese pages.
In the latter two years of the collaboration, communications, as indicated by a balance of questions and responses (see Table 2), seemed to indicate that communications flowed more smoothly possibly thanks to the fact that the English pages were created two weeks earlier in 2012 and 2013 than in 2011 -the first wiki page being created on November 6 during both years.A review of the numbers of comments and author responses shows that the percentage remained nearly equal for all of the wiki pages in 2012 and 2013 as in 2011 (see Table 3) in spite of a change in topic from posts about cultural topics to selfintroductory posts.While the percentages remained nearly the same, the total number of comments dropped across all categories -this decrease in comments invites further exploration through surveys or interviews of participants.
Logistical factors influencing communications success.Tanaka-Ellis (2011) compiled a comprehensive list of possible influencing factors for participant comments, including multiple topics in single comments, lack of implicitness, syntactic or pragmatic errors, or a change of topic by one of the participants (p.395).In reviewing the participants' comments, there was no indication that any of these reasons caused the lack of responses to questions or caused a decrease in the number of comments posted.The questions left unanswered were not linguistically difficult.Examples of unanswered questions from the English pages are: When is the best season for this kind of apple?, What is your major?, and What kind of videogames [sic] do you like to play?Examples from the Japanese pages include: あなたは来年シカゴのどんなところに行きたいですか？ [What part of Chicago do you want to go to next year?],なにをかきますか？[What do you draw?], and 何か英語で読み やすいおすすめの本があったら教えてください[Please tell me about easy-to-read English books you like].Instead of question or language difficulty, it would seem that the logistical factors mentioned above had a great deal of influence on the success of the intercultural communications in the 2011 wiki pages -92% of the questions unanswered questions were posed after December 10th, which was after the uwrf semester had ended (see Table 4).To determine the causes of the lack of answers to questions posed in the 2012 and 2013 collaborations, a review of each page author's final posts on their wiki page was conducted.On the 10 Japanese-language wiki pages with unanswered questions, six pages had zero or one comments by the author prior to the posting of questions by page visitors.On the remaining four pages, the unanswered questions were either posed by the author to visitors, posted too late for comment, or were ignored by the author.In reviewing the 29 English-language pages with unanswered questions over the three-year course of the collaboration, 17 of the pages had zero comments posted by the page authors.Five of the pages had questions posed after the official end of the collaboration in the second week of December; and three of the pages had questions posed by the authors to visitors who did not return to answer them.On the remaining four pages, the questions were posted within the timeframe of the collaboration but after the author's final comment, so there was no response given.
In light of this data, it is apparent that a key factor in ensuring successful communications and intercultural collaboration is active participation of all members of the project so that a complete dialogue among the members can be facilitated within the allotted timeframe.In order to support such active collaboration, the wiki page activities and comments must be an integral part of the participants' curriculum and syllabus (Brandl, 2012).The reasons for questions being left unanswered do not seem to be a result of the communication hindrances listed by Tanaka-Ellis ( 2011), but rather a result of logistical matters such as failure of authors to post any comments or the posting of questions after the collaboration's allotted timeframe had ended.With these factors in mind, the intercultural communications were successful among those participants of this project who did attend to the comments and questions of page visitors and remained engaged in the collaboration from the beginning to the end.

Analysis of participant comments
A review of the literature regarding the use of wikis in online intercultural collaborations showed that little research has been done on the types of participant comments and how the collaboration project topics affected the participants' interactions.This study reviews the types of comments made by participants in their l1 and l2 across the three-year term of this collaboration.The comments and responses were sorted into three categories based on their content and some comments or posts were included in two or three categories if their content warranted.The first category (A) was interpersonal communications; the second (B) was content-related communications; and the third category (C) was l2 languagerelated comments.Category A comments included statements or questions about personal likes and dislikes, preferences, or personal experiences, such as I like to eat that too.or Who is your favorite composer?Category B comments included statements or questions about cultural topics such as The hamburgers and pizza look good and I want to try eating them.or What is famous about your hometown?Category C comments included corrections or recasts of language errors, as in You wrote that you see American tv dramas, but we say, "watch tv dramas."Statements about participants' language ability, such as Your Japanese is very good.were not included in Category C. A discussion of the types of comments observed in this collaboration follows.
As might be expected, the change in the focus of the assigned topics for the collaboration from 2011 to 2012 and 2013 (see Table 1) resulted in a change in the type of comments posted.The topic of the 2011 wiki pages was culturally focused, thereby resulting in the majority of the comments relating to discussions about the cultural content on both the Japanese-and English-language wiki pages (see Table 5, overleaf).
As shown in Table 5, in 2011, the vast majority of the comments posted on Japanese pages and on the English-language pages were Category B, or culturally-related comments.The cause of this trend was most likely that in 2011, the wiki page topics were primarily culturally focused (see Table 6, overleaf).To clarify the topics in Table 6, Takamarukun is the city mascot used in the pictures of the cultural areas and Neputa and Nebuta are major local festivals.In 2012 and 2013, the focus of the assigned wiki page topics shifted to personal introductions to facilitate more interpersonal communications among the participants.As a result of this, the percentage of Category A comments increased dramatically on both the Japanese-language and English-language wiki pages (see Table 5).The percentage of the increase was greater on the English-language wiki pages; this is likely due to the logistical factors mentioned above which caused there to be more questions posed on the Englishlanguage pages -the fact that the American participants created their pages weeks ahead of the Japanese students and were waiting to begin the collaboration as soon as the Japanese students created their pages.Once the majority of the exchanges began on the Englishlanguage pages, their momentum kept the focus more heavily on the English-language page comments and interactions.
Comment examples from wiki pages.The interpersonal comments spanned a wide range of topics from introductions and statements of likes and dislikes, to hobbies, explanations of family members and hometowns, and school subjects and majors studied.Some examples of the types of interpersonal comments observed include: That reminds me of things I'd do with my family years ago.
I also like K-pop.What are your favorite bands?
私もダンスが好きです。 [I like dance too.] 私もフランス語をべんきょうしていました。 [I have also studied French.] The culturally-related comments covered a wide variety of topics, such as music, local and regional tourism information, traditional crafts and hobbies, and agricultural products, to name a few.Examples of the culturally-related comments observed in the interactions include: I have never heard of these musicians.Are they J-Pop artists?I don't listen to J-Pop but I would like to try it.
I highly recommend you come to Japan to see sakura in spring!!!You also be able to experience ohanami.Errors in the participants' comments.Looking across all three years of the collaboration, there were only two comments which dealt with language corrections, one of which was posted by an instructor.This was not because the participants' l2 skills were exceptionally high -many errors were evident in the comments and wiki page contents.The types of errors seen in the comments included missing English articles, incorrect word usage (My best favorite group is tvxq.),l1 transfer (I want [you to come to] Hirosaki Gakuin University badly.), and missing particles in the Japanese comments.Overall, the errors did not seem to inhibit communications or alter meanings.The instructors told the participants that language improvement was not the focus of this collaboration project, the focus was on learning about the l2 people and culture through intercultural exchanges.The students followed these instructions and made only one correction -a visitor to an English-language page noticed a spelling error in their own understanding of the abbreviated name of the Japanese university from HiruGaku to HiroGaku.Perhaps it can be assumed that the errors were not serious enough to affect the communications and understanding of the participants' posts, but this survey supports the findings of Brandl (2012), Lee (2011), andTanaka-Ellis (2011) that students rarely correct one another's linguistic errors within such collaborations.

Discussion
This study supported the findings of Chu et al. (2013), Loh andKim (2012), Miyazoe andAnderson (2010), andNielsen (2013) in demonstrating that wiki pages provide an excellent platform for online intercultural collaborations.Even with the technological inexperience that Murray and Blythe (2011) reported among Japanese university students, with explicit instruction and demonstration, the students were able to participate freely in the exchanges with the American students.The wiki pages enabled all participants to share pictures, links to other pages, and videos with one another as they shared personal and cultural information with their group members.
Regarding the types of comments seen, the fact that the content of the comments changed along with the focus of the assigned content of the wiki pages from 2011 to 2012 demonstrates that the participants will tailor their comments and inquiries to match the focus of an online collaboration project.However, even with a collaboration topic assigned, the participants will also engage in a variety of discussions about both interpersonal exchanges of information as well as about the cultural topic at hand.This study demonstrated that online intercultural collaboration projects provide language learners with an opportunity to learn not only about the target language culture, but also about the people who live in that culture.The participants of this study learned about the differences in university life for students in Japan and America, shared information about popular musical groups in each country, told each other about local places of interest to visit in the future, and discussed traditional cultural hobbies and activities.Participants of such intercultural collaborations are clearly able to broaden their understanding of other cultures by noticing differences and similarities in the l2 culture and their own.
In support of the findings of Brandl (2012) and Tanaka-Ellis (2011), this study found that participants in online intercultural collaborations do not engage in language correction or provide specific language-related feedback.As mentioned above, the only language-related feedback given in the three years of this study were a comment by an instructor and a minor correction to the spelling of the university's name.In spite of numerous errors on both the English and Japanese pages, they were not deemed to be significant enough to hinder communications or require correction.

Lessons learned
Throughout the three years of this study's online intercultural collaboration, the instructors have made slight changes to improve the students' experience and to make the collaboration more successful.Some of the problems with the implementation of the project were that the Japanese students were less technologically experienced than expected for university students.The American students had few problems with the technology, but the Japanese students required guidance and explicit, step-by-step instructions to perform the activities within this collaboration.In future collaborations, it would be beneficial for the instructor to integrate the collaboration project into the curriculum and to begin by taking the students to a computer lab to help them create their initial wiki pages which they can then improve upon in their own time.Relying on the students to create their pages on their own resulted in the delayed commencement of interactions seen during the first two years of this project.
Using small groups for the participants' interactions proved to be an important step in steering the students' interactions with one another, just as Oliveira et al. (2011) and Wang (2011) reported.Also, Brandl's (2012) suggestion to use open-ended interaction topics encouraged the participants to follow the tangents in their interactions like a normal conversation would.In choosing a topic for the focus of the an online collaboration, the data from this study as explained above showed that more interpersonal communications occurred when the page topics and discussion were centered around personal introductions and the sharing of personal information.When cultural topics were focused on, there were a large number of comments but fewer responses to inquiries, so it would be preferable to choose more personal topics for the intercultural collaborations because the participants seemed to have a stronger personal connection to the interpersonal communications instead of simply reporting on a cultural topic.Further, the asynchronous nature of wiki page-based interactions allowed the participants to consider their comments and responses more fully.This observation was in keeping with the findings of Hirotani and Lyddon (2013), Ito (2011), Nielsen (2013), and Tanaka-Ellis (2011).
Finally, even though Nielsen (2013) found that teacher participation in the interactions had little effect on the participants' interactions, Oliveira et al. (2011) found that the instructor's presence served to energize communications among groups which are slow to begin or are having difficulty working together.In this project, the instructors were often the first to comment on the students' pages, and these comments did spur some interactions among the participants, however, there were also instances in which the authors failed to respond or comment in spite of the teachers' presence.Overall, the instructors should be engaged in the collaboration project and monitor the participants activities so that they can intercede when activity decreases.

Conclusion
The online intercultural collaboration project conducted between Hirosaki Gakuin University and the University of Wisconsin -River Falls in 2011, 2012, and 2013 was a successful collaboration which allowed the participants to exchange both personal and cultural information via student-created wiki pages.To answer research question a), the participants did attend to previously posted comments when time allowed.The participants' exchanges were the most interactive when the students began the collaboration early in their fall semester; those who joined the collaboration late ran out of time to respond to questions posed by the other participants.This answers research question b) as well: the primary causes for questions going unanswered in this collaboration was a lack of time and failure of the participant to return to their personal wiki page after originally posting it.Given this information, the instructors must ensure that they prepare the students to begin the collaboration early, and integrate the project into the existing curriculum.The participants' interactions should also be monitored by the instructor to minimize the possibility of communications fading out in the middle of the collaboration.
In answer to research questions c) and d), this study demonstrated that the focal topic for online intercultural collaborations has an impact on the types of comments made by the students: when the topic was culturally focused, the participants' comments primarily related to cultural information; when the topic was personal, there was greater interpersonal engagement among the participants.Teachers should bear this in mind when planning an online intercultural collaboration, as tailoring the focus topic will allow the communications to remain relevant to the classroom curriculum as the instructor desires.
Finally, in preparing to conduct an online intercultural collaboration, instructors should remember that students may require more explicit instruction in the technical aspects of the collaboration -not all students are technologically savvy or comfortable interacting online.The instructor may need to provide individualized instruction at the beginning and during the interactions to ensure that all participants have a successful, educational, and enjoyable online intercultural collaboration project.
Future topics of research into online intercultural collaborations include a review of participants' comment length to determine whether a difference exists between participants comments in their l1 and their l2.Also, the number of comments can be reviewed to explore whether or not a preference is demonstrated for participants to comment in one language over another.Finally, a continuation of the research of Chu et al. (2013) can delve into student perceptions of participating in online intercultural collaborations using wikis.It is hoped that this study's data will continue to provide useful information for language educators who hope to incorporate such projects into their own curricula.

acknowledgement
The author would like to thank Professor Magara Maeda, Japanese instructor at the University of Wisconsin -River Falls, for her continued support of our online intercultural collaboration, as well as for her assistance in with this paper.Her enthusiasm and dedication to providing an excellent experience for our students has been inspiring.

Figure 2 .
Figure 2. Example of student's self-introduction wiki pages.

Table 1 :
Numbers of participating students and collaboration tasks

Table 2 :
Numbers of questions asked, and answered or unanswered within the wiki comments

Table 3 :
Total numbers of comments to wiki pages and the number of those comments which were responses by the authors

Table 4 :
Number of unanswered questions which were posted too late for response

Table 5 :
Categories of participants' comments in each year of the collaboration

Table 6 :
2011 Online intercultural collaboration wiki page topics.
Do you know ohanami?