3.1 Challenges associated with computer-mediated communication

Notwithstanding an interactionist approach being a regular part of face-to-face learning environments, it is difficult to implement such a methodology with isolated students without the use of technology. Indeed, Guldberg and Pilkington (Reference Guldberg and Pilkington2007) note how overcoming the distance has been a real challenge, but argue that modern day mobile technology now provides a possible means for promoting collaborative interaction between individuals studying in remote locations through the use of CMC (Gabriel, Reference Gabriel2004; Murphy & Coleman, Reference Murphy and Coleman2004; Paran, Furneaux & Sumner, Reference Paran, Furneaux and Sumner2004; Tudini, Reference Tudini2005; Weasenforth et al., Reference Weasenforth, Meloni and Biesenbach-Lucas2005). Other challenges associated with CMC indicate that distance learning courses can be impersonal, superficial, misdirected, potentially dehumanizing and depressing, and they can also disrupt the interactions that create a productive learning community (Rovai & Barnum, Reference Rovai and Barnum2003). Negative feelings of exclusivity, discouragement and frustration can also arise, thereby resulting in a sterile and impersonal form of communication (Murphy & Coleman, Reference Murphy and Coleman2004). Furthermore, despite the potential for the medium to support more reflection, knowledge construction and critical thinking, these benefits are not necessarily guaranteed (ibid.). For example, Pawan, Paulus, Yalcin and Chang (Reference Pawan, Paulus, Yalcin and Chang2003) highlight how students were found to engage primarily in serial monologues without instructors’ explicit guidance and teaching presence (see also Card & Horton, Reference Card and Horton2000; Wang & Woo, Reference Wang and Woo2007). Additionally, the time commitment necessary to participate in CMC activities can also be problematic (Gabriel, Reference Gabriel2004; Meyer, Reference Meyer2003; Paran et al., Reference Paran, Furneaux and Sumner2004; Wiesenberg & Hutton, Reference Wiesenberg and Hutton1996). As can be seen, numerous challenges are associated with CMC; however, numerous advantages have also been identified. Some of the more pertinent are described below.

3.2 Advantages associated with computer-mediated communication

A plethora of CALL studies have examined the interaction of students engaged in CMC while “chatting” (in a written form) using a computer as the communication instrument in collaborative activities (Beauvois & Eledge, Reference Beauvois and Eledge1996; Blake, Reference Blake2000; González-Lloret, Reference González-Lloret2002; Kern, Reference Kern1995; Kitade, Reference Kitade2000; Warschauer, Reference Warschauer1996). With access to the Internet, students are no longer bound by time or place (Beldarrain, Reference Beldarrain2006; Murphy & Coleman, Reference Murphy and Coleman2004); there is now the option to break down what Moore (Reference Moore1991) terms the “transitional distance” between those who are interacting to overcome the possibly detrimental effects of separation (including spatial, temporal, social, cultural, situational and/or psychological) from their instructor(s) and/or peer(s) (Kanuka, Rourke & Laflamme, Reference Kanuka, Rourke and Laflamme2007). For example, research shows how students all have an equal chance to participate when studying online (Birch & Volkov, Reference Birch and Volkov2007; Hemphill & Hemphill, Reference Hemphill and Hemphill2007; Kumari, Reference Kumari2001; Ortega, Reference Ortega1997) due to what Murphy and Coleman (Reference Murphy and Coleman2004:1) call the “equalizing effect” of CMC with regard to issues such as race, gender, accent or status (see also Warschauer, Reference Warschauer1997). In effect, a bridge is built (Tudini, Reference Tudini2005) which enables students to negotiate meaning as occurs in face-to-face oral interaction. Such a learning environment encourages noticing of errors, negotiations and modified output, thereby providing an optimal setting for second language acquisition (SLA). Importantly, research shows that students can learn equally well in online and face-to-face learning environments (Aragon, Johnson & Shaik, Reference Aragon, Johnson and Shaik2002; Moore & Thompson, Reference Moore and Thompson1990; Verduin & Clark, Reference Verduin and Clark1991). Furthermore, when compared to face-to-face discussions in the classroom, certain research actually indicates that online discussions can lead to greater perceived learning and higher levels of learner satisfaction (Rovai & Barnum, Reference Rovai and Barnum2003), and higher levels of quality and quantity of the content. CMC has also been shown to afford students numerous opportunities for interaction and communication whether undertaken synchronously or asynchronously (Gabriel, Reference Gabriel2004; Lock, Reference Lock2002; Turcotte & Laferrière, Reference Turcotte and Laferrière2004). Students can also provide feedback to one another and clarify their ideas (Weasenforth, Biesenbach-Lucas & Meloni Reference Weasenforth, Biesenbach-Lucas and Meloni2002; Wilson & Stacey, Reference Wilson and Stacey2004), thereby providing support in a collaborative manner.

3.3 Faults and remediation

Foreign or second language learning can be problematic in many areas, and situations frequently arise in which some form of remediation is necessary. Expanding on Corder’s original (Reference Corder1967) work, James (Reference James1998) explains how identifying the origin of any faults arising from these problematic areas helps to determine the degree of remediation that may be necessary. Accordingly, faults can be sub-divided into the following three main categories which are listed in increasing degree of severity: slips, mistakes and errors. In response to these faults, three approaches can be used to foster the development of self-correction strategies by students (Ros I Solé & Truman, Reference Ros, Solé and Truman2005). Listed in increasing order of the degree of remediation that is necessary, the approaches are: feedback, correction and remediation. However, focusing specifically on studies involving computer-assisted language learning, the term feedback is typically used in a more general sense to refer to information that is supplied to students regarding their work. Focussing on this area in more detail, the following section comprises a description of the most common types of computer-mediated feedback.

5.1 Participants

The 425 participants were first-year university English as a foreign language (EFL) majors in Japan. Classes were streamed according to the Kanda English Proficiency Test (KEPT) (Bonk & Ockey, Reference Bonk and Ockey2003). However, for the purpose of this study, students were assigned to one of two levels: “higher level” or “lower level”. The web-based reading exercises were completed by all of the fifteen classes in the year group. As students worked in pairs with classmates, proficiency levels within pairings were homogeneous. Materials developed for this study were trialled with 86 of the 425 students, leaving 340 students for the main study. However, records for 72 of the remaining students were omitted from the data analysis for reasons such as absenteeism, tardiness, odd numbers, technical difficulties and scoring 100% on the first comprehension exercise on the first attempt (these students did not benefit from feedback to promote comprehension which was the focus of this research). The main study and the statistical analysis were therefore performed on the data from the remaining 267 students.

5.2 Materials: reading materials

Materials used in this study comprised one reading text (see Appendix 1 for an excerpt) and two multiple-choice comprehension exercises, each with six questions (see Appendix 2 for an example question from Comprehension Exercise 1 and Appendix 3 for an example question from Comprehension Exercise 2). Therefore, the maximum score on each exercise was six points. While the questions in the two comprehension exercises were different, the same content points were covered by corresponding questions. In addition to the text and comprehension questions, three rounds of hints for the Elaborative feedback were also written. All materials formed part of a password-protected website.

5.3 Materials: feedback treatment

Murphy’s (2007) model shows how reading comprehension can be promoted through the introduction of computer-mediated Elaborative feedback and interaction with a partner. Using this model, the current research also focuses on both computer-mediated feedback (Elaborative and KCR) and pair work; however, students communicate via written CMC rather than face to face. Details follow below.

5.4 Procedure

The procedure used in this research is described below (see Table 1 for the descriptive statistics):

Table 1

Descriptive Statistics

1. (a) Students were divided into two levels of English proficiency. English proficiency level was thus the first independent variable.

2. (b) Type of feedback was the second independent variable; dyads were randomly chosen to receive either (i) KCR feedback only or (ii) up to three rounds of Elaborative feedback before KCR feedback. For the former condition, correct answers to all the comprehension questions were displayed when students submitted their answers to the first comprehension exercise for the first time. However, for the latter condition, computer-mediated Elaborative feedback was provided as follows:

   • • Having answered all questions and checked the answers on the first comprehension exercise for the first time, the first round of Elaborative feedback should encourage students to identify and correct incorrect answers by themselves. Therefore, Elaborative feedback is generated to direct students back to a key area in the text, for example: Please have a look at paragraph 6 again.

   • • If incorrect answers exist the second time students submit their answers for checking, Elaborative feedback should include a rephrased version of the question to help students identify and correct any mistakes/errors.

   • • If errors persist the third time students submit their answers, Elaborative feedback should include key information, sentences and/or phrases from the text, possibly in a paraphrased form.

   • • Finally, if errors persist following the fourth check of answers, the correct answers should be provided in the form of KCR feedback, and students should consult their teacher for further remediation.

It should be noted that the opportunity for random guessing is minimised by not initially identifying incorrect answers by question number. Instead, students are encouraged to read the feedback, re-engage with the materials, interact further with a partner, make any corrections as appropriate, and then recheck their answers. Furthermore, students are not supplied with the correct answers until they either receive the KCR feedback after the three rounds of Elaborative feedback or until they answer every question correctly.

1. (c) When participating in computer-mediated activities outside the classroom, numerous challenges can arise. These challenges may be related to studying without a teacher present, finding motivated partners at the right level for CMC, dealing with computer compatibility difficulties, overcoming any Internet availability, speed or bandwidth issues, and so on. In a conscious attempt to control for these factors so that any differences in comprehension scores could be attributed to the different types of computer-mediated feedback offered, this research was conducted in a computer-equipped classroom. However, in an attempt to also simulate remote locations, students were physically separated in the classrooms from their randomly assigned anonymous partners (Uribe et al., Reference Uribe, Klein and Sullivan2003).

2. (d) The first 25 minutes of the lesson for each of the fifteen classes was used for an introduction to the activity, which was given in both spoken English and written Japanese. All introductions were given by the researcher to maintain consistency. All students agreed to participate in the lesson. After listening to the instructions, students logged onto the online materials and then sent a message saying “Hello” to their partner on the pair’s pre-assigned bulletin board. After the introduction, students were given 50 minutes to read the text and complete the first multiple-choice comprehension exercise while interacting and collaborating with their partner through written messages using CMC. Students were next given a further ten minutes to complete the second comprehension exercise individually.

3. (e) To investigate the effects of ‘English proficiency level’ and ‘Type of feedback’ on all the dyads’ comprehension of the text during the first exercise, all students then individually completed the second comprehension exercise with KCR feedback only. The score on this second exercise was the dependent variable. All input data were stored in a database and analyzed quantitatively. Furthermore, transcripts of the interactions between students resulting from the CMC were then analyzed qualitatively. The results and their implications are discussed below.