This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.
This chapter describes the research methods for the study. It introduces the rationale of this study, linguistic target, research questions, participants, and procedures used for collecting data, and data analysis, which present a suitable statistical technique for the study, in more detail.
The rationale of the research
The importance of CF to Saudi students provides the rationale for the present study. From my point of view, using CF in teaching English has not been paid enough attention to in the Saudi educational system (SES).
As both a student and a teacher of English, I have often pondered over strategies and techniques regarding the most effective way of providing feedback to students, and whether or not feedback has an effective role in the way students acquire a certain linguistic target.
The main goal of choosing recasts (implicit error correction) which are considered to be the most generally used type of CF as suggested by Lyster (2007) and metalinguistic feedback (explicit error correction) which is considered to be the most efficient and least used type of CF as suggested by Ellis et al. (2006), is to look at how these types of CF interact with one of the most complicated areas of English grammar, which is the English articles.
The English article system is considered one of the most difficult parts of English grammar for non-native speakers and one of the latest to be completely acquired (Master, 1990). As many L2 learners of English, in spite of their level, face many difficulties in using English articles in addition, they may find it difficult to recover from these errors, especially for Arabic speaking learners of English in general (Noor, 1996; Sarko, 2008. English articles were chosen by the researcher to be tested in this present study. The researcher's purpose was not to look at whether CF assists the learning of a completely new structure, but whether it enables learners to gain greater control over a structure they have partially mastered. Usually, English teachers tend to pay no attention to students' English article errors as they may not affect comprehension (Master, 1995). Nevertheless, students have to master them as errors related to English articles may demonstrate that L2 learners have insufficient control over the language.
The study was conducted on 50 (40 experimental participants and 10 controls) Saudi female third-grade intermediate school students (aged between 14 and 15 years) who have had three years of learning English (beginners). Each twenty experimental participants were chosen randomly from two different third-grade classes (20 students from class A and 20 students from class B) and the 10 controls were chosen randomly from the two classes (five students from each class); also, they were chosen according to their scores in the English course in the last two semesters. In order to have a constant variation among participants, the researcher chose students whose marks varied between 90 and 95, 80 and 85 and 70 and 75. The rationale behind choosing them from different classes was that the researcher wanted the participants to be as varied as possible. One of the experimental groups (20 students) received recasts and the other (20 students) received metalinguistic feedback.
The reason behind conducting the research on third-grade intermediate school students was that the students at this stage might face difficulties in using English articles. This could be attributed to the fact that English articles do not attract a great deal of attention on the part of textbook authors and teachers in Saudi Arabia. On the other hand, the rationale behind choosing female students is that female researchers do not have access to male schools in Saudi Arabia.
The researcher made sure that all of the participants are of Saudi origin. Moreover, she asked for the school's assistance in order to overcome this issue by providing her with some information about their backgrounds. Moreover, the researcher asked each participant about whether or not she has attended any English courses outside the school, as the researcher wanted the participants to have a similar English level.
The researchers paid close attention to ethical issues throughout the study. As the study was conducted in an intermediate school, a clear approval was obtained from all the students, and they were told that their names would remain confidential. Moreover, the school's permission was sought. In addition, all participants were told before the task about the nature of the research and why the researcher has chosen to study this topic.
Data collection method
The data of this study have been obtained by one instrument which is test. In the following, the types of tests in this study will be presented.
Types of tests
The English articles rules  that were covered in this study as follows:
The is used with singular/plural/uncountable/abstract nouns when the reader/hearer knows what we are referring to.
A/an is used with singular or abstract nouns when the reader/hearer does not know what we are referring to.
Ø is used with plural/uncountable/abstract nouns when the reader/hearer does not know what we are referring to.
The researcher used two types of test. The idea behind examine the students on more than one test is that the task variance also adds support to the claim that L2 learners' performance must be investigated not only through a single measure because learner performance is significantly influenced by the type of tasks as suggested by (Tarone and Parrish, 1988).
Fill-in-the-Blank tests are commonly used from the classroom level to far larger scales to measure peoples' proficiency at English as a second language because of two reasons: (a) it is easy for test-takers to input answers and (b) computers could be used to mark them, hence marking is constant and objective (Sumita E. et al., 2005).
The fill-in-the blank test in this study consisted of around 32 sentences with 46 blanks and the students filled in the blanks with the appropriate articles. The researcher put some distracters in this test in order to prevent the students from becoming testwise.
Cloze test is a type of fill-in-the-blank test constructed by randomly deleting words from a passage, and replacing them with blanks (Aitken, 1977). Cloze tests are valid, reliable language tests that may simply be used by any teacher (Aitken, 1975).
The cloze test which is used in the current study included two passages with around 34 blanks and the students were asked to fill the blanks with the appropriate English articles. In fact, the researcher believe that, by not giving students choices and leaving them to fill the blanks with words that may make the sentence correct, could make the test difficult.
This test was managed because English articles sometimes behave differently according to the context. For example, the, in some cases, precedes a noun that has been previously mentioned, which is not usually the case with Arabic articles. Therefore, the researcher wants to see if the students' performance varies according to this. However, although there are some limitations with those two types of tests such as that they may facilitate cheating, the researcher followed the necessary procedures that prevented the students from doing this.
In fact, the reason behind choosing these types of tests is that other types such as oral story descriptions and grammaticality judgment tests may not be appropriate for them as they are not used to such tests which may impose further difficulty for them. Others such as translation tests may cause the students to be more influenced by their L1 (Arabic) which may lead them to committing more errors.
The tests administered over approximately one month and a half, and followed the pre-test, post-test and delayed post-test design.
All the participants, including the controls, were given the fill-in-the-blank and cloze tests. Since the number of the students is large, the researcher asked help from the English teacher at the school to invigilate the test with her. The duration of the test was around 70 minutes, which means that the students were given around one minute to answer each test item.
After the pre-test, the researcher marked the papers to see the nature of the errors so that she knew the materials that were necessary to be provided in the feedback sessions. The researcher found that most of the students' errors stemmed from their L1 affect. (Campillo, 2003)
The researcher planned to provide the students with a ninety-minute feedback session for three consecutive days.
On the first day, the recast and metalinguistic feedback groups separately received instructions on the usage of English articles in order to provide the students with the attention they deserve. On the second and third days, the recast group and metalinguistic group were given exercises to practice in class. In fact, they were given these exercises and provided with feedback at different times of day and in different classes. Each student received feedback with regard to her errors. An example of how feedback were communicated is, if a student from the recast group chooses the article the in [I like (a/an, the, ø) John], the researcher will say, 'The correct answer is [I like ø John]'  . But in the MG, the student will be told that the answer [I like ø John] because proper nouns in English are not normally preceded by any article as is the case in Arabic. That is, the metalinguistic group was provided with reasons why their answers were wrong, whereas the other group was not. In fact, the researcher put into account that for CF to be effective, it should be: a) noticed; b) salient; and c) directed towards a single linguistic feature that the learner is ready to acquire and this is what was suggested by Sheen (2008).
The immediate post-test was completed on the day following the last feedback session, and the delayed post-test  was completed after around one month and a half  in order to see the long-term effects of CF. All the students, control, recast, and metalinguistic groups, were given the same fill-in-the-blank and cloze tests to see how the feedback had influenced their performance.
The fill-in-the-blank test was scored
As the purpose of the research was to look at the number of errors the participants committed before and after they were given feedback, and to see to what extent they would recover from these errors, the researcher used a quantitative method in the current study. The rationale behind choosing such a way is that it helps to make the results less affected by the researcher's subjectivity and characteristics. Moreover, quantitative research helps the researcher 'generalize the findings beyond the cases that make up the sample' (Bryman, 2004: 77).
In quantitative research, the researcher usually wants to be able to suggest that the results are not particular to the participants in the study. Thus, other researchers may easily check the results and the procedures of the study associated with quantitative research. For these reasons, the validity  and reliability  of the research will be enhanced.
All of the statistical calculations in the present research were carried out using SPSS and Excel 2007. A one-way Analysis of Variance (ANOVA) will be carried out on the three groups (recast, metalinguistic and control groups) to see if there are significant differences between the pre-tests, post-tests and delayed post-tests of every group. Then the researcher will perform a one-way ANOVA on the three pre-tests, post-tests and delayed post-tests of each group to see if there are significant differences between the groups on each test. Moreover, since there are two types of tests: fill in the blank and cloze tests, the researcher will do the same thing on each type of test, to see if each group's performance varies from one type of test to another.
After this, the researcher will perform a Paired Samples T Test to see whether, for example, the recast group fill in the blank pre-test, post-test, and delayed post-test's performance varies from their performance on the cloze-test pre-test post-test, and delayed post-test.
Aitken, K. G. (1975) 'Problems in Cloze Testing Re-Examined.' TESL Reporter 8.
Aitken, K. G. (1977) 'Using Cloze Procedure as an Overall Language Proficiency Test.' TESOL Quarterly, 11, 59-67.
Bickerton, D. (1981) Roots of Language. Ann Arbor: Karoma publishers.
Bryman, A. (2004) Social Research Methods. Oxford Oxford University Press.
Campillo, P. (2003) 'An Analysis of Implicict and Explicit Feedback on Grammatical Accuracy.' Journal of English and American Studies, 209-228.
Ellis, R., Loewen, S. and Erlam, R. (2006) 'Implicit and Explicit Corrective Feedback and the Acquisition of L2 Grammar.' Studies in second language acquisition, 28, 339-368.
Hawkins, R. (2001) Second Language Syntax: A Generative Introduction. Oxford: Blackwell.
Lyster, R. (2007) Learning and Teaching Languages through Content: A Counterbalanced Approachlearning and Teaching Languages through Content: A Counterbalanced Approach. Amsterdam: John Benjamins Publishing Co.
Master, P. (1990) 'Teaching the English Articles as a Binary System.' TESOL Quarterly, 461-478.
Master, P. (1995) 'Consciousness Raising and Article Pedagogy.' In Belcher, D. and Brain, G.(eds) Academic Writing in a Second Language. Norwood, NJ: Ablex.
Sheen, Y. (2008) 'Recasts, Language Anxiety, Modified Output, and L2 Learning.' Language Learning, 58, 835-874.
Sumita E., F., S. and S., a. Y. (2005) 'Measuring Non-Native Speakers' Proficiency of English Using a Test with Automatically-Generated Fill-in-the-Blank Questions.', Proceedings of the Second Workshop on Building Educational Applications using Natural Language Processing. Ann Arbor, US, Measuring Non-Native Speakers' Proficiency of English Using a Test with Automatically-Generated Fill-in-the-Blank Questions.: pp. 61-68.
Tarone, E. and Parrish, B. (1988) 'Task-Related Variation in Interlanguage: The Case of Articles.' Language Learning, 38, 21-44.
Yates, S. J. (2004) Doing Social Science Research. London: SAGE Publications Ltd.