3 Diversity in learner training
Introduction
Learner training for computer-assisted language learning (CALL) is a process aimed at promoting the development of technology competence specifically for the purpose of second language acquisition.1 As evidence of its growing recognition in language teaching, the recent framework document for the TESOL technology standards (Healey et al., Reference Healey, Hegelheimer, Hubbard, Ioannou-Georgiou, Kessler and Ware2009) lays out a set of performance indicators for language learners, with the implicit assumption that teachers are responsible for training the students to achieve them. Beyond that general competence, though, is the more specific competence needed to use technology successfully within a given environment, task, or software program. Since this is a more widely, though still quite inadequately, studied area, it will be the focus of the remainder of this chapter. It should be noted, however, that many of the points discussed here are also relevant for developing general competence for using technology in language learning.
In line with the theme of this edited volume, the purpose of the present chapter is to explore diversity in learner training for CALL. Specifically, we will look at two distinct dimensions: diversity in the training process and diversity in the individuals and groups undergoing the training. By combining these two in a single work, we hope to achieve the goal of providing a broad overview of the possibilities for learner training along with a description of some of the issues that remain to be resolved in accommodating the diversity inherent in groups and individuals. From our experiences with CALL learner training over the past few years, it is clear that both the collective and the individual perspectives need to be addressed to improve the effectiveness of this endeavor.
Overview and general issues
Diversity in the learner training process
Background
Although a few articles previously mentioned the impor tance of training language learners for general technology competence (e.g. Beller-Kenner, Reference Beller-Kenner, Egbert and Hanson-Smith1999), Barrette (Reference Barrette2001) was perhaps the first to recognize the need for training those already considered “computer literate” – university students in the US. Her paper began with a focused review of fourteen CALL research articles appearing in the CALICO Journal and Foreign Language Annals from 1997 and 1998, noting that there was little explicit evidence either of participants’ previous computer literacy or of training them with the applications being studied. At the outset of her own study, she surveyed her students’ familiarity and comfort level with nine applications ranging from word processing to building web pages. She then provided technical training in the use of these applications along with tasks for employing them in foreign language learning contexts. Her results showed impressive gains in students’ self-reported comfort levels with these applications, and she concluded that profiling students’ computer literacy and providing training are important steps in effective use of CALL. More recently, Winke and Goertler (Reference Winke and Goertler2008b) conducted a wide-ranging survey expanding on Barrette’s work, involving 911 first- and second-year foreign language students at Michigan State University. They discovered that many students did not have access to or proficiency with the specialized tools needed for CALL.
Hubbard (Reference Hubbard, Fotos and Browne2004) approached learner training for CALL explicitly, suggesting that it should take a prominent place in the field. That paper offered five practice-based principles for CALL learner training consistent with much of the literature on other types of learner training. Kolaitis and a group of colleagues (Kolaitis et al., Reference Kolaitis, Mahoney, Pomann, Hubbard, Hubbard and Levy2006) conducted a three-year project based on applying Hubbard’s model to a set of courses within a community college ESL program and concluded that it not only influenced student interaction with CALL materials positively but also had a noticeable washback effect on the participating instructors’ classroom teaching. Adopting the same model, O’Bryan (Reference O’Bryan2008) demonstrated that even a small amount of training (three ten-minute sessions) led to measurable, though not quite statistically significant, distinctions in student use of help options in online reading materials, as well as clearly increased awareness of their value.
A review of subject characteristics in CALL literature across seventy-eight research studies in four major technology and language learning journals from 2000–2004 noted that only 31 percent of those studies included any kind of learner training (Hubbard, Reference Hubbard2005). When training was provided, it was typically only basic training at the beginning of the study to insure that students knew how the application functioned. Evidence of research involving any training on how to link the application or activity to language learning objectives or of training beyond the beginning stage of the study was rare. Yet, somewhat surprisingly, twenty-two of the studies (29 percent) mentioned in the discussion section that additional training was warranted and could have led to more favorable results. More interestingly, fourteen of these had not mentioned any training prior to the study and three others had explicitly specified that no training was provided as part of the research design.
The literature to date thus suggests that it is common practice to offer little if any learner training before turning students loose on a CALL software application, task, or activity. However, it also suggests that providing such training can result in improved performance. In the remainder of this section, we explore the options in doing so.
Three-part framework
This section addresses diversity by looking at the different types of learner training available to a teacher or developer. It reviews a three-part training model (Romeo and Hubbard, Reference Romeo, Hubbard, Levy, Blin, Siskin and Takeuchi2010) distinguishing technical, strategic, and pedagogical training, and discusses issues in the type of training that may be useful depending on the proficiency level and readiness for self-directed learning of the language students.
For students to be effective users of a computer tool or learning application, they must first understand how to operate it and then become comfortable with its operation. Although there is often a general sense that today’s “digital natives” (Prensky, Reference Prensky2001) need no such training, there is evidence to the contrary. As noted previously, Barrette (Reference Barrette2001) and Winke and Goertler (Reference Winke and Goertler2008b) have provided strong support for the contention that many students are not sufficiently prepared to engage in effective use of computers for language learning and that technical training is therefore needed. Such training is not limited to basic “computer literacy” but crucially encompasses three additional areas. One is the use of general applications, such as audio and video recording, speed controls in media players, and advanced searching with Google and other search engines, that many learners may not have proficiency in from their non-CALL uses of computers. A second is the use of language-specific applications, such as keyboarding skills necessary for foreign character sets. The third and perhaps most overlooked is an understanding of the options and controls in a particular dedicated CALL application. Even if the courseware itself offers tutorials and help options, as many commercial products do, students may forget or fail to discover useful functions provided by the program without sufficient training (Kolaitis et al., Reference Kolaitis, Mahoney, Pomann, Hubbard, Hubbard and Levy2006).
While technical issues are specific to CALL learner training, the concept of strategy training is already familiar to most in the language teaching community. In fact, strategic training for CALL is an extension of the strategy instruction that blossomed in the 1980s as exemplified in the work of Oxford (Reference Oxford1990) and others. A certain portion of CALL strategic training involves strategies analogous to those in non-CALL environments, such as preparing for listening or reading by previewing and reflecting on the likely content of a text. Others, however, are more specific to CALL, such as opening a separate text window for a transcript of a web-based video and placing it beside the video for easy reference (Figure 3.1). To clarify, to know how to open, move, and resize multiple windows on a computer desktop is a technical skill; to use that knowledge deliberately to aid comprehension and vocabulary development when needed is a strategy.
Figure 3.1 Using multiple windows to facilitate listening comprehension and vocabulary development
It is well established that there are differences in the effective use of various support features in CALL applications, such as clicking on hypertext links in digital readings to get L1 translations, L2 definitions, graphic support, or cultural notes (e.g. Chun and Plass, Reference Chun and Plass1996) and responding to computer-generated feedback (e.g. Pujolà, Reference Pujolà2001; Heift, Reference Heift2002). These differences in effective use could be lessened if more learner training infusing such features strategically were provided.
A full discussion of learner strategies available for CALL is beyond the scope of this chapter, although it is clear that there are ones that transfer from non-CALL settings as well as those specific to CALL settings (e.g. monitoring the ongoing transcript during an online chat, slowing recorded speech using media player controls, and hypertext linking to comprehension supports of various types in an online reading). It is also clear that there is potential for training that impacts strategy use of the four widely cited types: cognitive, metacognitive, social, and affective strategies (Oxford, Reference Oxford1990). In particular, in online settings where the learner is more independent than in a traditional class, expansion of metacognitive strategies seems warranted.
The final type of learner training, pedagogical, involves giving learners some of the knowledge and skills provided in the education of a language teacher. An example of pedagogical training would be to introduce students to the concepts and supporting research underlying the distinctions between top-down, bottom-up, and interactive processing in listening and reading (Peterson, Reference Peterson and Celce-Murcia2001). This is an extension of the principle from Hubbard (Reference Hubbard, Fotos and Browne2004) “Give learners teacher training,” and it is based on the assumption that CALL environments generally provide learners with more choices and thus more potential control over their learning paths than other environments. Therefore, the decisions a learner makes when using CALL materials and tools should be informed by an adequate knowledge base. Pedagogical training is sometimes integrated with strategic training, providing the foundation for students to determine when and how to use known strategies as well as to devise new ones.
To summarize, technical training provides learners with expertise in how to use the options and controls of both general and specific applications on the computer for language learning purposes. Strategic training focuses on what to do to support certain task and learning objectives, including how to link sequences of strategies or techniques into coherent procedures. Pedagogical training provides the foundation for helping learners determine specific learning objectives and understand why to use certain techniques and procedures to achieve those objectives. The relative weight given to these three provides a range of diversity for environments with learner training.
Determining an effective balance of these training areas in a given setting is a challenge. It is not enough to focus on technical training alone with the tacit assumption that once students understand the operation of the application that they are using, they will be able to use it appropriately for their language learning purposes. It is important to integrate strategic training so that learners can make the connection between what the technology has to offer and how those affordances can be manipulated in an effective manner. The area of pedagogical training is somewhat more problematic. As we will see in the next section, there is an issue of time: all training takes time away from other learning activities, and pedagogical training in particular may seem far enough removed from the goals of the class to not warrant that additional time. Further, when cultural diversity is considered, there may be resistance among students from certain groups to taking on a responsibility that they believe should belong to the teacher. Finally, there is the question of diversity in motivation to consider: just because students are provided with foundational knowledge about language learning does not mean that they will be able to connect that knowledge to selecting and using strategies effectively. These factors need to be taken into account when deciding what the proper balance among the three types of training should be for a given class or individual.
Timing and other options
In addition to the diversity provided by the three types of learner training above, there are several other dimensions of diversity in the training process to consider. These include the timing of the training, the form of the training, and the intensity of the training.
Timing. In settings where learner training has been provided, there is a tendency to begin with a preliminary tutorial session and then follow it by allowing the learners to proceed on their own with little if any additional assistance. This may work well in situations where the training needed is relatively simple or when the training builds on areas already familiar to the students. In many cases, however, it is unlikely that a single round of training will have the continued impact, especially before students have familiarized themselves with a particular application or learning task. One of the learner training principles from Hubbard (Reference Hubbard, Fotos and Browne2004) is to make training cyclical, so that technical information, strategies, and pedagogical generalizations are recycled and expanded upon in a natural spiral. An extension of that principle is that both technical and strategic training may be more efficient if learners are first given the opportunity to explore a new application or learning environment. In fact, rather than overwhelming students with information at the beginning in a single intensive tutorial session, they may be encouraged to explore the CALL application or environment on their own. Xie (Reference Xie2002), for example, showed how allowing students to use a Chinese chat program anonymously and casually first helped them build familiarity with operating the character sets before engaging in target language activities that would be assessed. It is also possible to allow use of the first language to familiarize students with computer-mediated communication (CMC) activities and novel learning tools such as concordancers. Finally, an alternative to the timing in the types of training above, or even a natural complement, is to take an interventionist approach. In this case, training is offered when problems are noted, either at the request of the learner or because of observations from the instructor.
Form. There is also diversity in the form of training. As with other areas of language instruction, teachers may take either an inductive or deductive approach to learner training, or a combination of the two. In an inductive or exploratory approach, learners are presented with an opportunity to discover an application’s controls, help options, and so on along with useful paths through the task or activity, either on their own or collaboratively with peers. This is then followed by checking to see if their interactions are in line with the teacher’s expectations, followed by hints or other forms of guidance if they are not. In a deductive approach to learner training, students are provided with clear instructions, sometimes through modeling, and then given either structured practice or the opportunity to directly make use of the instruction in using the CALL application or engaging in the CALL task. Because of their more organized structure, deductive approaches are often seen as being more efficient than inductive ones. However, they are not necessarily as memorable or engaging.
Regardless of whether one takes an inductive or deductive approach or some combination of the two, a related question is whether students are internalizing the training primarily on an individual level or whether collaborative learning is occurring. In inductive approaches in particular, the ability to discuss one’s insights and interactions with the CALL application or activity with peers can have a positive effect. In lab situations, this can be supported by having students work in pairs in front of a single computer. In settings where students work independently, individual experiences can be posted onto a discussion board or class time can be taken for small group collaborative debriefings (Hubbard, Reference Hubbard, Fotos and Browne2004).
Intensity. As was the case historically with strategy training in areas such as reading and listening, one of the barriers facing teachers who want to incorporate CALL learner training into their classes is that it takes time. Another area where diversity comes into play, then, is determining how much time a teacher or a student can spend working independently to provide sufficient benefit, without impacting too much on other class and homework activities. As O’Bryan (Reference O’Bryan2008) showed, even a small amount of time could have a measurable impact in student interaction with materials and awareness of learning supports provided by the software. At the other end of the spectrum is what we have called “pervasive” learner training (Romeo and Hubbard, Reference Romeo, Hubbard, Levy, Blin, Siskin and Takeuchi2010), where learner training is integrated into virtually every class session and students demonstrate through reports and individual meetings what they have and have not clearly understood with respect to the training.
Diversity in the realm of intensity is not solely a factor of available time but is also related to diversity issues in the learners. Some learners are prepared to make profitable use of the fruits of training, while others may consider training time to be of little value. Kolaitis et al. (Reference Kolaitis, Mahoney, Pomann, Hubbard, Hubbard and Levy2006) noted that providing CALL training where learners were expected to reflect on their learning processes did not work as well for students with lower language proficiency. Boling and Soo (Reference Boling, Soo, Egbert and Hanson-Smith1999) and Brandl (Reference Brandl1995) similarly note that advanced students are able to make more effective use of the range of controls in language learning software than those with lower proficiency. Research on cognitive resource limitations suggests an answer, namely that less complex and more familiar strategies would be easier for students at lower levels of language proficiency to implement because their attention is focused so heavily on conscious manipulation of the language itself.
To summarize, we have seen that there are a number of dimensions that can be controlled to yield a range of diversity in learner training processes. These include selecting a balance among the three main types of training – technical, strategic, and pedagogical – as well as determining the timing, form, and intensity of the training. Given the limited research to date, it is not yet possible to identify what an optimal training process would be. However, recognizing the range of options, reflecting on the potential costs and benefits of various combinations of alternatives, and collecting data on what does and does not yield desired results, should allow for continuing progress to be made in understanding what is worthwhile. In order to understand the process more fully, however, we need to take into account the diversity of the learners themselves at both the group and individual levels.
Learner diversity
Much of language teaching methodology and second language acquisition (SLA) research aims at capturing similarities. In the case of teaching, this is for pragmatic reasons because teachers work with collections of individuals and in many cases with collections of sub-groups as well, as happens in the multicultural English as a second language (ESL) classrooms common to intensive English programs. Similarly, for much SLA research, a primary objective is to capture the common elements of second language learning so as to characterize its connections to cognitive processes and social interaction adequately. However, both teachers and learners have long recognized that there are differences across individuals and groups in language learning, and that understanding these variations can be critical in determining what leads to successful language learning. Various aspects of diversity in learners with respect to technology are covered elsewhere in this volume – see Chapters 2 and 4. Here we concentrate on viewing that diversity from the perspective of the process and progress of learner training.
Skehan (Reference Skehan1989) was perhaps the first to offer a book-length overview of language learner diversity. He identifies four SLA models relevant to the topic: the monitor model (Krashen, Reference Krashen1981), the good language learner model, Carroll’s model of school learning, and a disjunctive model. For most of his work, Skehan builds on the Naiman et al. (Reference Naiman, Frohlich, Stern and Todesco1978) “good language learner” model, giving particular attention to the construct of aptitude, which he says “still generates the most consistent correlations with language learning success” (p. 8). Since Skehan, others have looked at various individual or group factors in an attempt to account for degrees of success and failure of language learning. Lightbown and Spada (Reference Lightbown and Spada2006), for instance, devote an entire chapter to the topic. Among other factors they mention differences in learner beliefs (though mainly at the group rather than the individual level) as important, especially where those beliefs are in conflict with the teaching methodology, even when that methodology may represent “best practices” elsewhere. Another area of diversity is gender: Oxford and Nyikos (Reference Oxford and Nyikos1989, p. 295), for example, note “profoundly significant sex differences in strategy choice” in three of the five factors they studied in 1,200 university-level language students.
A relatively comprehensive list of major categories of diversity can be drawn from the keywords in the chapter titles of Griffith’s (2008) edited volume, Lessons from Good Language Learners. Contributions to this volume cite research that variation across particular categories separates the “good language learner” from the rest. The categories mentioned there under the label of “learner variables,” which can also be seen as a checklist for evaluating diversity for our purposes, include the following:
motivation
age
learning style
personality
gender
strategies (strategy use)
metacognition
beliefs
culture
aptitude.
Complementing these characteristics, a primary area of diversity in language classrooms is that of second language proficiency. Some classes are designed for beginners, some for intermediate students, and some for advanced, with various gradations in between. Proficiency has a connection with some of the elements above: more advanced learners, for example, tend to be more autonomous (Boling and Soo, Reference Boling, Soo, Egbert and Hanson-Smith1999), engage in more metacognition, and use more strategies than beginners.
Diversity can also be seen in the way individuals and groups interact with computer-based material or with one another in online tasks. A few examples are illustrative of the range. Heift (Reference Heift2002) noted that students using an intelligent tutoring system for German varied in using the correct answer feedback provided by the program. Some checked the correct answer before inputting anything (browsers), some checked the answer after attempting and getting an error message (peekers), and others (18 percent) never looked at the correct answer but continued trying to get it on their own (adamants). Grgurovic and Hegelheimer (Reference Grgurovic and Hegelheimer2007) studied students who were given the opportunity to select either transcripts or subtitles as help options in a listening activity. When surveyed at the end, the majority preferred subtitles, but five of eighteen preferred transcripts. Looking at small groups rather than individuals in collaborative tandem-learning sessions with students from the US and Germany, Belz (Reference Belz2001) showed diversity in terms of high- and low-group functionality. She identified several factors including differences in language level, cultural expectations, and access to the Internet that influenced the relative success or failure of the groups.
Finally, in any learning environment employing computers, differences in underlying technical proficiency provide another important area of diversity. This can have an influence at the whole class level, where all or most of the students may have limited expertise in some area, or at an individual level, where a specific student or students may be foundering due to lack of the needed technical skill and knowledge, perhaps because of a general technophobia (Robb, Reference Robb, Hubbard and Levy2006). As mentioned previously, Winke and Goertler (Reference Winke and Goertler2008b) provide a wealth of data demonstrating that contemporary university language students are not, despite their “digital nativeness,” prepared to utilize digital technology effectively for language learning. Students may also differ greatly in their overall “eLearning” experience and proficiency, an area that falls between general technology proficiency and proficiency specific to CALL applications and tasks.
Example
Diversity in an ESL listening class
As an example of some of the preceding concepts, we look at a study of learner training in an advanced ESL listening class at Stanford University taught by one of the authors in the fall of 2008. The course was built around online audio and video materials for both whole class and individual assignments, so it was important for students to be able to use them effectively. As this course was the final one in our listening sequence, some attention was also paid to learner training to provide students with the skills and knowledge for continuing language development on their own once the course ended. The overall results have been reported on in detail elsewhere (Romeo and Hubbard, Reference Romeo, Hubbard, Levy, Blin, Siskin and Takeuchi2010) – here we concentrate on those that relate to issues in diversity discussed above.
The fifteen students from two sections were from Korea, China, Taiwan, Japan, and the Philippines, with the majority in their third quarter of full-time graduate study at Stanford. There were three parts to the course: in-class listening practice, discussion, and learner training; class homework; and individual projects. The projects, explicitly included to accommodate individual diversity, were negotiated with each student to allow them to pick objectives and material types in line with their needs and interests. They were required to do a minimum of two hours per week, ideally divided into three forty-minute sessions, though they were allowed flexibility as long as they met the minimum total time. They submitted weekly reports specifying the objective, materials, time spent, procedures, and reflective comments. Data for the overall study were collected from pre-/post-surveys of technology proficiency, pre-/post-listening tests, notes, and videos from individual tutorial sessions (five per student), weekly student reports, class notes, and an end of course interview. The data reported below are primarily from the class notes and student reports, which turned out to have the richest and most detailed material.
In terms of diversity in the training process, all three types of learner training (technical, strategic, and pedagogical) were integrated into the course and the intensity of the learner training was pervasive. Training on various aspects of computer-based listening was introduced in each class, sometimes deductively by advance demonstration, and sometimes inductively by asking students to try on their own to figure out an effective way of meeting a particular objective and then discussing it subsequently in small groups or individual tutorials. Key techniques and strategies were reviewed regularly both in class and in the tutorial sessions.
The course introduced a number of strategies for improving listening comprehension, and these strategies were often supported by pedagogical training as described above. The goal was to move students towards a state of autonomy where they would be able to make informed decisions regarding the learning objectives of each homework session, the materials used, and the strategies chosen in working with them. Some of the more prominent technology- focused strategies were pausing and rewinding the media, slowing the playback speed, setting the graphic equalizer on the media player to be optimized for speech, expanding the player size for greater control, using subtitles and online transcripts, linking to online dictionaries, doing background research for pre-listening (e.g. linking to Wikipedia), and listening to materials multiple times with different objectives for each pass. Other techniques were introduced for students to work specifically to build their language proficiency, such as creating personal vocabulary lists, reviewing vocabulary with an online program using spaced repetition, written dictation to link form to meaning and improve chunking in processing, and oral summarization to connect listening and speaking.2 Reflecting on the learning experience to enhance metacognitive awareness was encouraged in class, the weekly reports, and individual meetings.
Our goal was to observe how much the techniques helped students improve their listening, or at least positively change the way they interacted with online listening materials. On the surface, the students were relatively homogeneous, coming from similar educational and language learning backgrounds in Asia, all having proficiency advanced enough to be admitted to full-time graduate study (most in engineering or science) and all showing high levels of general computer proficiency. The results for the group overall were quite positive (see Romeo and Hubbard, Reference Romeo, Hubbard, Levy, Blin, Siskin and Takeuchi2010), and we have continued to develop the training component of the course.
However, it was clear that there was also diversity among the learners in a number of areas, especially with respect to the independent listening project that was a hallmark of the course. For this project, which lasted for seven weeks during the ten-week quarter, students determined which online materials they would use, what their learning objectives were, what strategies and procedures they would use to accomplish those objectives, and how they would time the sessions. Below, we discuss a few of the more prominent areas where diversity was observed.
Scheduling was one major area of diversity among learners. Some students arranged a schedule and stuck closely to it, others attempted a schedule but often failed to adhere to it, while a few claimed that sticking with a schedule did not fit their graduate lifestyle, preferring to work on their projects after all the homework for their major classes was done (often on weekends). During week seven, they also experimented with changes in the frequency and length of listening sessions, after the instructor specifically asked them to attempt study sessions of twenty minutes (one each day for six days) and compare the experience with the previously recommended three sessions of forty minutes each. The majority reported that they preferred shorter, more frequent sessions, because they felt that they learned more and were able to form better study habits. However, there were three students (20 percent) who felt that twenty minutes was just too short to cover the material in any satisfactory way, and two of the other students actually offered that six longer study sessions of forty minutes each would have been optimal.
Diversity among students could be seen quite clearly in students’ goals. There were many cases of individuals who were ultimately looking for ways to improve their spoken English and would therefore gravitate to media which contained naturalistic native speech, such as TV shows. However, others who were more focused on listening comprehension, especially as it related to academic English, found that news and other informational media was a much more useful resource. One example of this was an interesting contrast which was noted between Chuck and Joe (pseudonyms are provided to preserve student anonymity). Chuck reported experimenting on a variety of strategies such as listening to one clip multiple times (multiple listening), reading the transcript, researching background information, shadowing (see below), and dictation. He focused on vocabulary many times in his reports and reported moving from entertainment TV shows to online news because he found the latter more effective for his comprehension goals. Joe reported experimenting with most of the same techniques, except for pre-listening research, but added written summarization as a comprehension check and an opportunity to engage in language production. In the end he reported preferring to use entertainment to informational media, linked to his interest in everyday conversational English. For these two students, their individual goals played an important role in determining what strategies and materials were effective for them.
While some subjects were quite adventurous with their projects, most of them stuck with a relatively small set of favored strategies, although these sometimes changed in response to discussions during the individual meetings. Besides the strategies presented to students, other techniques were already familiar to them from prior study overseas, but often they had not questioned their effectiveness. An example of one such popular strategy was “shadowing,” which entails listening to pre-recorded media (news, or some other entertainment media) and simply following along, repeating the words and sentences as they are heard. Despite its popularity, particularly in Korea and Japan, its effectiveness has not been well documented. However, Tamai (Reference Tamai2002) found that a program which focused on shadowing produced a significant positive effect in overall listening scores, though not in other areas such as vocabulary and word span with numbers. Two students in particular explored this technique in their weekly listening activities. Over the course of several weeks, Nathan reported a greater awareness of the processes involved in shadowing, and, while he did not stop using it, did use it less frequently. More interestingly, he tried adding dictation as a way of increasing his involvement with the media following pedagogical training that explained the relationship between dictation and awareness of language forms.
Nathan’s case can perhaps shed some light on students’ gravitation to certain strategies. He had a new MP3 player and was in the habit of downloading podcasts and listening to them during the day. His personal situation also had an influence on this study technique; he lived with his wife and new-born daughter and reported difficulty finding time to focus on listening outside of the early morning hours. In the end, he found that increasing the frequency of study times while reducing the length was the most productive plan for him. This shows how some characteristics of group diversity develop: aspects of personal situations and commonalities in available equipment can lead to a natural grouping. Although an instructor could dismiss such trends, saying familial status or purchasing habits should not be allowed to influence academic achievement, it could be argued that finding ways to accommodate such diversity is a more effective means of reaching a larger number of students.
During the training process, students seemed to go through three stages, although we did not have a sufficient number of subjects to make a strong case for this claim. First, they began to realize exactly what made certain media difficult. Many were not exactly aware that characteristics such as the speech rate, content familiarity, and personal interest could be the source of not being able to understand. This realization was reflected in early reports and meetings of almost every subject. At the next stage, they were able to implement strategies to make listening easier and interact more with the material. Finally, these strategies were internalized and used when the need arose, often without a lot of conscious effort, for example, toggling subtitles on and off as needed relative to the task objective rather than leaving them on or off throughout the whole listening activity. This last step is important and students showed diversity here – some failed to demonstrate in their reports and tutorial sessions that they had reached that last stage. An example of one who apparently had was Sam. He reported reviewing media multiple times, recognizing the value in a second or third listening, and was observed hiding subtitles when they were not needed to keep them from distracting him. However, when asked about techniques he consciously used to improve listening comprehension or proficiency, he did not include these, presumably because he no longer considered them “special.”
Finally, the topic of subtitles was a very important one for many subjects, especially given the widespread view among both teachers and students that they are more of a hindrance than a help to developing comprehension skills and increasing proficiency. Based on their experience both in class and with their projects, many concluded that listening without them at least once but then using them to confirm meaning and to identify new vocabulary was the most effective strategy. One interesting case was Oliver, who found that the subtitles in the media he was using were not accurate, so he abandoned using them altogether. Previous research in the use of subtitles has largely concluded that they can serve a purpose both for improving comprehension and building vocabulary (Danan, Reference Danan2004) and even for increasing the amount and quality of subsequent speech (Borras and Lafayette, Reference Borras and Lafayette1994). However, others caution against their use, arguing that reliance on them may hamper comprehension of real-world speech (Vandergrift, Reference Vandergrift2004). Our findings indicate that, rather than ignoring them or using them constantly, it would serve both instructors and students well to identify methods of using (and not using) them to scaffold learning. Clearly, this is an area ripe for more targeted research.
Implications
Essentially, the main implication of this chapter is that to the extent that learner training for CALL is worthwhile, it is itself a form of teaching, and like other teaching it can be done with more or less effectiveness depending on a variety of factors. These include not only diversity in the students’ readiness and motivation, but also how well the training curriculum fits the students and is executed by the instructor. We have also noted that training takes time away from other class and homework activities that instructors may feel are equally or even more critical to the success of the course.
Our research on learner training in an advanced listening class has led to recognizing some areas for further study and development relevant to the theme of diversity. As noted, the learner training process used in the example class can be characterized as pervasive, with significant elements of all three types of training: technical, strategic, and pedagogical. To a large extent, students were encouraged to experiment with a variety of materials, strategies, and procedures to find those that suited them best. These students were given a lot of independence fairly quickly. Even considering their relatively advanced language proficiency and maturity compared to most language students, for some that independence may have been excessive. They might have profited from a more guided approach, perhaps led by a set of integrated strategies and procedures representing a sort of “best practice.” Once they had become familiar with this foundation, they might have found it much easier to select from among those candidates and then adjust them to fit their individual needs.
Given the fact that we are well into the third decade of CALL as a professional field with its own dedicated conferences and journals, learner training for CALL is a remarkably understudied area. Nevertheless, the absence of a solid research base is not a suitable argument for ignoring it in practice. We have increasing evidence that learners do not, automatically, know how to utilize digital materials and tools effectively for language learning just because they have familiarity with other uses of computers. There is a tendency in any new domain to oversimplify it – our goal in this chapter has been to show that monolithic approaches to training, however elegant in appearance, will not accommodate the range of diversity discussed herein, and that further exploration is called for.
Conclusions
This chapter has discussed diversity in learner training for computer-assisted language learning applications across two major dimensions – diversity in the training process and diversity in the learners themselves – with an eye to understanding how the processes and the learner diversity can be reconciled effectively. Given the limited research to date, it is not possible to determine what an optimal process would be, either in general or for a specific group or individual. Additionally, though not a part of the present chapter, there is the rather critical question of what the actual content of the training should be. Determining that will require continued experimentation based on our best hypotheses of what is likely to work coupled with collection and analysis of learner data to help us confirm, reject, or refine those hypotheses. We can also predict with some confidence that diversity in both general and application-specific strategies in using technology for language learning will continue to grow as the technology changes.
Irrespective of these daunting challenges, CALL learner training is a very important topic for the future. As we continue taking early steps towards understanding and integrating it into language teaching, we need to recognize the rich range of diversity available in the training process in addition to the well-established diversity within individuals and groups of learners to which technology use only adds further variables.
1 The term “learner training” is used here because of its wide recognition in both CALL and general language teaching literature. One could argue that the goal often goes beyond training to “education,” implying the ability to make independent informed choices, rather than mechanically following directions. This is analogous to the shift from the term “teacher training” to “teacher education” in recent decades. In the present work we will continue with “learner training” for ease of exposition but the terminology is an issue which is worth revisiting as this subdomain of CALL – and language learning in general – grows. See Sinclair (Reference Sinclair2006) for an interesting discussion.
2 See the class notes at www.stanford.edu/~efs/693b-08 for additional details.
