Guenette, D. (2007). Is feedback pedagogically correct? Research design issues in studies of feedback on writing. Journal of Second Language Writing, 16, 40-53.
Rather than trying to interpret the conflicting results of research on feedback in L2 writing (form) as a demonstration of effectiveness or ineffectiveness, Guenette argues that such disparate results to date (which has failed to guide teachers in their practice) may be highly attributable to research design and methodology, as well as the "constellation of variables" that have been ignored. For example, in scrutinizing the research methodology, she points out that efficacy of feedback may be attributable to proficiency levels, which is a variable that is rarely measured and reported accurately. Also, while there is already a deficit in the amount of research that uses both a control group and experimental group, those few that do usually fail to keep all other variables consistent. Another design feature she raises attention to is the "elicitation task" (which I think refers to 'demand'--what do the students have to do with the feedback?). Some of the variables that teachers usually ignore include classroom contexts (is the control group and experimental group receiving the same instruction? Does their instruction place attention to focus on form?), student incentive (are students being graded on the way they react to the feedback?), and motivation, which is also usually overlooked. The article concludes by summarizing the benefits of descriptive studies, which reveal other dimensions of feedback: students' ability to engage with feedback, the type of errors that benefit from feedback, students' perceptions and preferences, and individual differences.
This study can be useful for researchers to fine-tune our research design and methodology so that our research can begin to become more comparable and meaningful; clearly something that is not really happening. We need to be more vigorous in our search for other variables that might affect studies, and this article serves as a good reminder for how to improve our research. I enjoyed the study because it's my area of interest and was vaguely familiar with many of the studies that were only briely referred to. So I'd recommend it as a reading....but maybe only if you're involved in research on feedback in L2 writing with a focus on form.
A study that I spotted in this article that seems particularly interesting to me was:
Han, Z. H. (2001). Fine-tuning corrective feedback. Foreign Language Annals, 34, 582-595. This is a longitudinal case study that reveals individual differences are an important factor in determining whether feedback is effective or not, and calls for fine-tuning the feedback to the learners' specific problems.
Wednesday, August 29, 2007
Evaluation of Loewen and Philip (2006) By Ping
Loewen, S., & Philip, J. (2006). Recasts in the adult L2 classroom: Characteristics, explicitness and effectiveness. Modern Language Journal, 90, 536-556.
This study examined the provision, the nature, and the effectiveness of recasts in 12 adult ESL classrooms. The findings indicated that recasts were widely used and were beneficial at least 50 % of the time. Although other forms of feedback had similar success rate measured by posttests, recasts differed in the connection between learner response and test performance. The study suggests that recasts vary widely in terms of stress, intonation, number of feedback moves, length of recasts, and degree of explicitness. These characteristics of recasts are likely to affect their effectiveness in classroom context. More specifically, it points out that stress, declarative intonation, and number of feedback moves can be associated with the effectiveness of recasts since these elemens increase the salience of recasts and therefore learners respond to them as explicit corrections. The study also suggests that other factors, such as degree of differences between the learner's utterance and the recast and the nature of discourse in which recasts are provided, should be taken into account.
I think this article brings up an important issue we talked about in previous session: the degree of explicitness of negative feedback. It investigates the differences within recasts and considers the impact of their characteristics on their effectiveness, instead of classifying them as one generic type of implicit negative feedback. Also, it highlights the importance to not only look at learners' immediate uptake, but also consider the overall interactional organization of the discourse. Using learners' immediate response to feedback might not be adequate to assess the effectiveness of recasts. For example, multiple feedback moves can contribute to further learner engagement and higher learner production, therefore serving as predictor of successful uptake.
If you are interested in variable elements of recasts and the complexities of their characteristics, this will be a good read. It has a comprehensive discussion on different perspectives regarding the benefits and limitations of recasts. I think it's a good introduction to those who are not familiar with recast studies. Even though I got confused with its coding categories, overall, I will recommend it to be added to the 750 reading packet.
This study examined the provision, the nature, and the effectiveness of recasts in 12 adult ESL classrooms. The findings indicated that recasts were widely used and were beneficial at least 50 % of the time. Although other forms of feedback had similar success rate measured by posttests, recasts differed in the connection between learner response and test performance. The study suggests that recasts vary widely in terms of stress, intonation, number of feedback moves, length of recasts, and degree of explicitness. These characteristics of recasts are likely to affect their effectiveness in classroom context. More specifically, it points out that stress, declarative intonation, and number of feedback moves can be associated with the effectiveness of recasts since these elemens increase the salience of recasts and therefore learners respond to them as explicit corrections. The study also suggests that other factors, such as degree of differences between the learner's utterance and the recast and the nature of discourse in which recasts are provided, should be taken into account.
I think this article brings up an important issue we talked about in previous session: the degree of explicitness of negative feedback. It investigates the differences within recasts and considers the impact of their characteristics on their effectiveness, instead of classifying them as one generic type of implicit negative feedback. Also, it highlights the importance to not only look at learners' immediate uptake, but also consider the overall interactional organization of the discourse. Using learners' immediate response to feedback might not be adequate to assess the effectiveness of recasts. For example, multiple feedback moves can contribute to further learner engagement and higher learner production, therefore serving as predictor of successful uptake.
If you are interested in variable elements of recasts and the complexities of their characteristics, this will be a good read. It has a comprehensive discussion on different perspectives regarding the benefits and limitations of recasts. I think it's a good introduction to those who are not familiar with recast studies. Even though I got confused with its coding categories, overall, I will recommend it to be added to the 750 reading packet.
Evaluation of Loewen and Erlam (2006) by BoSun Choi
Loewen and Erlam (2006). Corrective Feedback in the Chatroom: An experimental study Shawn. Computer Assisted Language Learning , 19, 1-14
This quasi-experimental study tried to investigate the effectiveness of the two types of corrective feedback, i.e., recasts (implicit) vs. meta-linguistic information (implicit) on in online chatting with 31 beginning L2 learners of English. The target structure is regular past tense (verb+ed), which is known as a morpheme acquired later in morphemes studies (see Dulay and Burt, 1974). After taking pretest, the participants went through 56 minutes of corrective feed session (treatment), where they received either type of corrective feedback while completing the two tasks; story retelling after seeing a picture with written narratives and verbal description of the pictures. Their learning is measured by timed and untimed grammaticality judgment test (GJT) immediately and two weeks later. ANOVA analysis displayed insignificant difference for the two corrective feedback types and for time, meaning that the two groups there is no significant gains in response to either feedback type nor was there significant gain over time.
Reading this article is somewhat useful in a sense that it gives some insight into differences between face-to-face and online study. The suggestion of three features of online chatting is acceptable to explain insignificant amount of uptake as follows; 1) reduced immediacy of the feedback due to the overlap between interlocutor turns 2) the lack /reduced incidence of uptake in response to feedback. 3) students’ frequent going off the target and instructor’s less control over the off-topic. However, such suggestion is a bit mitigated when the author mentions that the participants’ low proficiency level may lead to insignificant amount of uptake. If the participants were not ready to learn the target form their proficiency level would serve as a conclusive factor. Even though the picture is puzzling, the characteristics of online chatting suggested in this study would give some ideas for further studies about corrective errors in online setting, which may be beneficial those who are interested in CALL and error correction.
I personally enjoyed reading it since the research gave me some insights about the characteristics of online feedback. It would be more enjoyable, however, if this study provided more sound explanation about the design. First, it did not explicitly explain why it adopted timed and untimed GJT for measurement different from the one in Ellis et al.’s(2006) study, by which this research is motivated for replication. Even though the design of this study is not new but directly employed from classroom research, this article may be enjoyable for this class in a sense that it describes some characteristics of online classroom setting, which greatly differs from off-line class. Especially, the instructor’s low level of control over student’s utterance seems to be interesting factor in response to its effect on learning since CALL literature pointed out that student centeredness is one of the important benefit of the online instruction.
I would like to recommend this article for the class since there has been scarce literature about the effect of online feedback in SLA literature. Despite of the result that there was no significant effect for either online corrective feedback, it is worth reading the description of the characteristics of online chatting. Given students’ growing need for application of technology to language classrooms, it would be worth further studying the effect of using different tool in language classroom.
Evaluation of Ellis and Sheen (2006) by Kevin
Ellis and Sheen (2006) Reexamining the Role of Recasts in Second Language Acquisition. SSLA, 28, 575-600
This article assesses the current state of research concerning recasts. The authors identify and suggest solutions for several types of problems that have occurred in many studies to date. Specifically, they are argue that there are a variety of recast forms and functions and these are often not taken into account; the type of corrective strategy being studied, that is (implicit or explicit/negative or positive) is often unclear; social and sociocognitive perspectives are often not considered or factored into results; the significance of learner repair following recasts has not been examined in enough detail; developmental readiness should be investigated more thoroughly as a factor in acquisition by recast; other corrective strategies (such as elicitation, clarification requests, translation, metalinguistic feedback) have not received as much attention as recasts in the research literature.
One important issue that the article brings up is the question of classifying recasts. I think many of us tend to use the word recast to cover a huge group of varied corrective feedback. Should we classify types of recasts and treat them as separate types of one phenomenon for research purposes? The authors mention several types of recasts including repetition, reformulation, corrective and non-corrective, full or partial, single or multiple, simple complex and context-dependant. Clearly, the types of recasts listed above are not all equally effective and researchers should be aware of these distinctions when planning research. In addition to types of recasts, other factors also complicate the definition of a recast. Feedback can be in the form of addition, deletion, substitution or reordering. The student/teacher orientation must be taken into account in terms of whether the language is being treated as an object or for message conveyance; Is the recast didactically or communicatively motivated? The authors suggest the possibility of classifying in relation to effects recasts have on acquisition or successful uptake.
Another question that has been addressed, but not adequately, according to the researchers, is whether or not recasts are implicit or explicit forms of feedback. The authors state that it often depends on the linguistic or discourse context and that researchers should be more aware of this fact. Another interesting and related question is whether or not recasts result in metalinguistic awareness or an internalization of rules without being consciously aware of a rule or pattern. The research up to this point does not give a clear picture. Are implicit recasts noticed as being corrections or something else? The larger question posed here concerns the relationship between noticing and recasts. To what degree are the different types of recasts noticed? Is noticing a factor in acquisition by recasts? Added to this are the roles of positive and negative evidence and the further complicating effect that they have on the type of recast being used. The authors state tentatively that positive evidence arises naturally and may or may not be noticed and acquired and negative evidence is explicit feedback. The recast itself could be both, depending on the learner interpretation and the discourse context.
Having established that recasts come in many sizes shapes and forms, the authors address the most important question of all; Do recasts facilitate acquisition? That is, should we even bother?
The research to date shows that there is certainly potential, especially if the recast focuses on a single feature or is emphasized in some way. The research is inconclusive. A numbers of factors seem to contribute to the effectiveness of recasts for facilitating acquisition. The literacy or proficiency as well as the learner’s orientation may also affect recast effectiveness and should be taken into account when deciding on a recast strategy. Individual learner differences, type of recast, target of the recast, developmental readiness of the learner also appear to be part of the picture, but to what extent is not know at this point. Several questions remain; Is uptake and repair dependant on instructional and social context? . Is noticing present only if uptake and repair is present? Does subsequent usage of correct forms contribute to acquisition?
A lot of ideas are covered in this article, each one worthy of a long discussion. It may help to summarize some of the main ideas. This is done below:
1. Recasts come in a variety of forms and should not be treated as homogeneous
2. Recasts can be functionally different and this may affect acquisition.
3. Recast can be considered explicit as well as implicit feedback
4. Learner orientation is a factor in deciding if recasts can be considered positive or negative feedback.
5. Learners may not recognize corrective force of recasts
6. Socio-psychological factors, individual learner differences, target language and developmental readiness may determine receptiveness
7. The role of uptake is uncertain
8. Research tends to focus on focused recasts. These results cannot be applied to classroom situations where recasts are extensive.
9. No clear evidence that recasts work better for acquisition than other forms of interaction.
In terms of our previous discussion about effectiveness, the authors define an effective recast as one that facilitates acquisition. I recommend this article for inclusion in the reading packet.
This article assesses the current state of research concerning recasts. The authors identify and suggest solutions for several types of problems that have occurred in many studies to date. Specifically, they are argue that there are a variety of recast forms and functions and these are often not taken into account; the type of corrective strategy being studied, that is (implicit or explicit/negative or positive) is often unclear; social and sociocognitive perspectives are often not considered or factored into results; the significance of learner repair following recasts has not been examined in enough detail; developmental readiness should be investigated more thoroughly as a factor in acquisition by recast; other corrective strategies (such as elicitation, clarification requests, translation, metalinguistic feedback) have not received as much attention as recasts in the research literature.
One important issue that the article brings up is the question of classifying recasts. I think many of us tend to use the word recast to cover a huge group of varied corrective feedback. Should we classify types of recasts and treat them as separate types of one phenomenon for research purposes? The authors mention several types of recasts including repetition, reformulation, corrective and non-corrective, full or partial, single or multiple, simple complex and context-dependant. Clearly, the types of recasts listed above are not all equally effective and researchers should be aware of these distinctions when planning research. In addition to types of recasts, other factors also complicate the definition of a recast. Feedback can be in the form of addition, deletion, substitution or reordering. The student/teacher orientation must be taken into account in terms of whether the language is being treated as an object or for message conveyance; Is the recast didactically or communicatively motivated? The authors suggest the possibility of classifying in relation to effects recasts have on acquisition or successful uptake.
Another question that has been addressed, but not adequately, according to the researchers, is whether or not recasts are implicit or explicit forms of feedback. The authors state that it often depends on the linguistic or discourse context and that researchers should be more aware of this fact. Another interesting and related question is whether or not recasts result in metalinguistic awareness or an internalization of rules without being consciously aware of a rule or pattern. The research up to this point does not give a clear picture. Are implicit recasts noticed as being corrections or something else? The larger question posed here concerns the relationship between noticing and recasts. To what degree are the different types of recasts noticed? Is noticing a factor in acquisition by recasts? Added to this are the roles of positive and negative evidence and the further complicating effect that they have on the type of recast being used. The authors state tentatively that positive evidence arises naturally and may or may not be noticed and acquired and negative evidence is explicit feedback. The recast itself could be both, depending on the learner interpretation and the discourse context.
Having established that recasts come in many sizes shapes and forms, the authors address the most important question of all; Do recasts facilitate acquisition? That is, should we even bother?
The research to date shows that there is certainly potential, especially if the recast focuses on a single feature or is emphasized in some way. The research is inconclusive. A numbers of factors seem to contribute to the effectiveness of recasts for facilitating acquisition. The literacy or proficiency as well as the learner’s orientation may also affect recast effectiveness and should be taken into account when deciding on a recast strategy. Individual learner differences, type of recast, target of the recast, developmental readiness of the learner also appear to be part of the picture, but to what extent is not know at this point. Several questions remain; Is uptake and repair dependant on instructional and social context? . Is noticing present only if uptake and repair is present? Does subsequent usage of correct forms contribute to acquisition?
A lot of ideas are covered in this article, each one worthy of a long discussion. It may help to summarize some of the main ideas. This is done below:
1. Recasts come in a variety of forms and should not be treated as homogeneous
2. Recasts can be functionally different and this may affect acquisition.
3. Recast can be considered explicit as well as implicit feedback
4. Learner orientation is a factor in deciding if recasts can be considered positive or negative feedback.
5. Learners may not recognize corrective force of recasts
6. Socio-psychological factors, individual learner differences, target language and developmental readiness may determine receptiveness
7. The role of uptake is uncertain
8. Research tends to focus on focused recasts. These results cannot be applied to classroom situations where recasts are extensive.
9. No clear evidence that recasts work better for acquisition than other forms of interaction.
In terms of our previous discussion about effectiveness, the authors define an effective recast as one that facilitates acquisition. I recommend this article for inclusion in the reading packet.
Evaluation of Sachs & Polio (2007) by Hung-Tzu
Sachs, R., & Polio, C. (2007). Learners’ uses of two types of written feedback on a L2 writing revision task. Studies in Second Language Acquisition, 29, 67-100.
Reformulation refers to a native speaker's rewriting of an L2 learner's written composition in order to make the language seem more nativelike, while keeping the content of the original intact (Thornbury, 1997). Through three-stage composition-comparison-revision tasks, Sachs and Polio’s study (2007) examined first, the effectiveness of reformulations versus written error corrections; second, the relationship between higher quality noticing and revision outcome; and third, whether think aloud while comparing reformulated writings make a difference in linguistic accuracy of learner revision.
Fifteen adult learners of English participated in a repeated measure study with three experimental conditions: error correction, reformulation, and reformulation + think-aloud condition. Though this first experiment suggested that noticing of feedback was related to accuracy of subsequent revision, the error correction group outperformed the reformulation group probably because the error correction group had more time to memorize corrected forms. A second experiment was carried out to in attempt to eliminate the time variable. The nonrepeated measure design with 54 learners found similar results that reformulation did not prove to be more helpful than error corrections. Also, learners who were in the reformulation group outperformed their counterparts in the reformulation + think aloud group. Sachs and Polio suggested that while verbal protocols gave insights on learner-internal process in relation to written feedback, results on the effectiveness should be interpreted with care.
I do not think this article is the best option to be included in our course reading mainly because I doubt the feasibility of using reformulation as a feedback technique in writing courses. The problem lies not only in the time and effort required from the teachers to rewrite every student's composition but also the possibility that such technique could greatly endanger learners' ownership in writing. The study did not include any discussion on such concerns. In explaining the process of reformulation, the researchers indicated that grammatical errors, style, cohesion, and vocabulary were all part of the correction. This to me, is a very vague description on just how much rewriting was done.
While I feel that reformulation might not be the most practical written feedback in writing classrooms, I did enjoy the discussion on think-aloud as a measure of noticing. The researchers pointed out the difficulties in coding and determining awareness in metalinguistic verbalizations and also discussed the complex relationship between noticing during think aloud, accuracy in revision, and long-term acquisition.
The study might be useful for those considering using think-aloud as an effectiveness measure, but overall, my evaluation to the article is not recommended.
Reformulation refers to a native speaker's rewriting of an L2 learner's written composition in order to make the language seem more nativelike, while keeping the content of the original intact (Thornbury, 1997). Through three-stage composition-comparison-revision tasks, Sachs and Polio’s study (2007) examined first, the effectiveness of reformulations versus written error corrections; second, the relationship between higher quality noticing and revision outcome; and third, whether think aloud while comparing reformulated writings make a difference in linguistic accuracy of learner revision.
Fifteen adult learners of English participated in a repeated measure study with three experimental conditions: error correction, reformulation, and reformulation + think-aloud condition. Though this first experiment suggested that noticing of feedback was related to accuracy of subsequent revision, the error correction group outperformed the reformulation group probably because the error correction group had more time to memorize corrected forms. A second experiment was carried out to in attempt to eliminate the time variable. The nonrepeated measure design with 54 learners found similar results that reformulation did not prove to be more helpful than error corrections. Also, learners who were in the reformulation group outperformed their counterparts in the reformulation + think aloud group. Sachs and Polio suggested that while verbal protocols gave insights on learner-internal process in relation to written feedback, results on the effectiveness should be interpreted with care.
I do not think this article is the best option to be included in our course reading mainly because I doubt the feasibility of using reformulation as a feedback technique in writing courses. The problem lies not only in the time and effort required from the teachers to rewrite every student's composition but also the possibility that such technique could greatly endanger learners' ownership in writing. The study did not include any discussion on such concerns. In explaining the process of reformulation, the researchers indicated that grammatical errors, style, cohesion, and vocabulary were all part of the correction. This to me, is a very vague description on just how much rewriting was done.
While I feel that reformulation might not be the most practical written feedback in writing classrooms, I did enjoy the discussion on think-aloud as a measure of noticing. The researchers pointed out the difficulties in coding and determining awareness in metalinguistic verbalizations and also discussed the complex relationship between noticing during think aloud, accuracy in revision, and long-term acquisition.
The study might be useful for those considering using think-aloud as an effectiveness measure, but overall, my evaluation to the article is not recommended.
Evaluation of Ammar and Spada (2006)’ article by Yun Deok Choi
Ammar, A., & Spada, N. (2006). One size fits all? : Recasts, prompts, and L2 learning. Studies in Second Language Acquisition, 28, 543-574.
Based on the previous research findings that corrective feedback (CF) is facilitative for L2 learning, this quasi-experimental study investigated the effectiveness of CF and that of recasts and prompts, respectively, in conjunction with learner proficiency level. Sixty-four francophone students from an intensive ESL program were assigned into a recast, prompt, and control group. The target feature was third-person singular possessive determiners (PDs), namely, his and her, which are known to be difficult for French speaking ESL learners. As for treatment, instructional intervention was implemented, and it consisted of 1 instruction on PDs and 11 practice sessions. Learners’ knowledge of the target feature was measured by pretest, immediate posttest and delayed posttest that were composed of a passage correction and an oral picture description task. The researchers discovered the followings: utilizing CF during communicative activities was more beneficial for learners than without using it, prompts were more facilitative than recasts for learners with low proficiency, the two CF techniques were more helpful than no CF for learners with low proficiency, prompts were more beneficial than recasts for learners with low proficiency, and learners with high proficiency were able to get equal benefit from both CF techniques. The researchers explained the better efficiency of prompts by addressing their explicitness and numerous opportunities to uptake they provide.
It will be very helpful for my classmates to read this article because it provides concisely summarized information about the previous studies on the effectiveness of corrective feedback on L2 development. That is, if you read it, you will easily understand the core aspects of prior studies in relation to the current topic since it deals with the previous experimental studies as well as descriptive studies related to the topic in depth. In addition, it also offers sufficient details with respect to every section, so you can readily comprehend not only the whole experimental situation but also the subsequent results and explanations for them.
I personally enjoy reading this article very much since I am interested in both the effectiveness of corrective feedback like recasts and prompts in classroom settings and learner proficiency level, which is the topic of the article. In addition, it was written in a very well-organized manner with plenty of detailed explanations. For example, at the end of the article, the researchers posed limitations of the current research that we have to keep in mind when we estimate the research and they also suggested some interesting areas to be dealt with in the future study. This information helps readers to comprehend the whole story more completely. Therefore, I strongly recommend that we should add it to our reading packet.
Ammar, A., & Spada, N. (2006). One size fits all? : Recasts, prompts, and L2 learning. Studies in Second Language Acquisition, 28, 543-574.
Based on the previous research findings that corrective feedback (CF) is facilitative for L2 learning, this quasi-experimental study investigated the effectiveness of CF and that of recasts and prompts, respectively, in conjunction with learner proficiency level. Sixty-four francophone students from an intensive ESL program were assigned into a recast, prompt, and control group. The target feature was third-person singular possessive determiners (PDs), namely, his and her, which are known to be difficult for French speaking ESL learners. As for treatment, instructional intervention was implemented, and it consisted of 1 instruction on PDs and 11 practice sessions. Learners’ knowledge of the target feature was measured by pretest, immediate posttest and delayed posttest that were composed of a passage correction and an oral picture description task. The researchers discovered the followings: utilizing CF during communicative activities was more beneficial for learners than without using it, prompts were more facilitative than recasts for learners with low proficiency, the two CF techniques were more helpful than no CF for learners with low proficiency, prompts were more beneficial than recasts for learners with low proficiency, and learners with high proficiency were able to get equal benefit from both CF techniques. The researchers explained the better efficiency of prompts by addressing their explicitness and numerous opportunities to uptake they provide.
It will be very helpful for my classmates to read this article because it provides concisely summarized information about the previous studies on the effectiveness of corrective feedback on L2 development. That is, if you read it, you will easily understand the core aspects of prior studies in relation to the current topic since it deals with the previous experimental studies as well as descriptive studies related to the topic in depth. In addition, it also offers sufficient details with respect to every section, so you can readily comprehend not only the whole experimental situation but also the subsequent results and explanations for them.
I personally enjoy reading this article very much since I am interested in both the effectiveness of corrective feedback like recasts and prompts in classroom settings and learner proficiency level, which is the topic of the article. In addition, it was written in a very well-organized manner with plenty of detailed explanations. For example, at the end of the article, the researchers posed limitations of the current research that we have to keep in mind when we estimate the research and they also suggested some interesting areas to be dealt with in the future study. This information helps readers to comprehend the whole story more completely. Therefore, I strongly recommend that we should add it to our reading packet.
Subscribe to:
Posts (Atom)