THEORETICAL AND EMPIRICAL ISSUES IN THE STUDY OF WRITTEN CORRECTIVE FEEDBACK

Get Complete Project Material File(s) Now! »

The irregular past tense

The results for research question 1 indicate that only the two direct CF groups improved in accuracy significantly or near significantly between Task 1 and Task 2. From Task 1 to Task 2 all the remaining groups’ scores improved but not significantly. All of the groups’ scores deteriorated between Task 2 and Task 3. However, these differences were not statistically significant. The results for research question 2 show that there were no significant group differences in the gain scores in either the short-term or the long-term.
This was the case for both the individual groups and for the comparisons involving directness and focus.
It would appear on the face of it that that there is a short-term within-group effect in the case of focused and unfocused direct CF. However, in order to obtain a clearer picture of the effect of the different kinds of CF on accuracy in the use of irregular past tense forms, it is necessary to examine to what extent learners were able to use the specific irregular forms that they produced incorrectly and that were corrected in their subsequent writing.
This is because, as pointed out before, the acquisition of irregular verb forms involves item- rather than system-learning. For the four CF groups, the data were inspected to determine whether those irregular verbs that were corrected in Task 1 were not used, used correctly or used incorrectly in a
subsequent piece of writing. This was undertaken for both Task 2 and Task 3 (see Appendix E). Table 26 shows the frequency of learners’ correct and incorrect use of those irregular verb forms that had been corrected in Task 1. It is clear that by and large the learners were not able to make use of the corrections they had received in new pieces of writing. Less than 50% of the irregular verbs that had received correction were subsequently used correctly.

Investigating the distribution of the past tense scores

In line with Brown’s (1973) criteria, participants who scored above 90 % on Task 3, which operated as a pre-test, were removed from this part of the research as they were considered to have already acquired the structures. The distributions of the remaining scores were examined to see whether they were normally distributed. Kolmorgorov-Smirnov tests found that some groups were not normally distributed (refer Appendix C), so nonparametric rather than parametric tests were used. For the analysis of group differences, while there were no group differences evident in Task 3, it was decided to generate gain scores for the sake of a consistent analysis across all the research questions investigating group differences (i.e. RQ2, RQ4 and RQ6). The gain scores were generated by calculating the difference between the scores in Task 3 and the revised version of Task 3.

READ Influencing factors on online decision-making process

Number of corrections received

The number of corrections received in corrective feedback session 2 was tallied for each of the four groups for both the regular and irregular past tense. Table 29 presents the descriptive statistics for the regular past tense. A Kruskal-Wallis Test was computed to gauge whether there were any significant differences between the four CF groups in the number of corrections received. The result was not significant (χ² (3, 47) = 4.27, p = .23). The descriptive statistics for the number of corrections received by the groups for the irregular past tense are presented in Table 30. A Kruskal-Wallis Tests was also conducted on these scores to test whether there were any significant differences in the number of corrections received. Again the result was not significant (χ² (3, 35) = 3.84, p = .28).

CHAPTER ONE INTRODUCTION
1.1 Overview .
1.2 Educational content
1.3 Personal context
1.4 Theoretical context
1.5 Key issues in written corrective feedback research
1.6 Outline of thesis
CHAPTER TWO THEORETICAL AND EMPIRICAL ISSUES IN THE STUDY OF WRITTEN CORRECTIVE FEEDBACK .
2.1 Overview .
2.2 A framework for investigating oral and written corrective feedback
2.3 Theoretical perspectives on written corrective feedback .
2.4 An overview of research on written corrective feedback
2.5 Ellis’ typology of written corrective feedback .
2.6 The debate about the efficacy of written corrective feedback .
2.7 Issues of research methodology
2.8 State of current knowledge about written corrective feedback .
2.9 Gaps in the literature.
CHAPTER THREE PILOT STUDY
3.1 Overview
3.2 Research questions
3.3 Target structures
3.4 Population and sample
3.5 Design
3.6 Instruments
3.7 Procedures
3.8 Data collection
3.9 Analysis .
3.10 Results
3.11 Discussion
3.12 Conclusion
3.13 Changes to the main study
CHAPTER FOUR METHODOLOGY
4.1 Scope of the research
4.2 Research questions
4.3 Target structure
4.4 Population and sample
4.5 Design
4.6 Data collection
4.7 Procedures
4.8 Data collection schedule
4.9 Analysis .
CHAPTER FIVE THE EFFECT OF CORRECTIVE FEEDBACK ON NEW PIECES OF WRITING
5.1 Overview
5.2 Investigating the distribution of past tense scores
5.3 Number of corrections received
5.4 Results for the regular past tense .
5.5 Results for the irregular past tense
5.6 Summary of statistically significant results
5.7 Discussion .
5.8 Summary
CHAPTER SIX THE EFFECT OF CORRECTIVE FEEDBACK ON A REVISED VERSION OF A TEXT
CHAPTER SEVEN THE EFFECT OF CORRECTIVE FEEDBACK PLUS REVISION ON NEW PIECES OF WRITING
CHAPTER EIGHT CONCLUSION