Sequential Mixed-Method Design Research Paper

This study will use explanatory sequential mixed-method design – using two methods in a study sequentially. First will be a quantitative design; secondly, a qualitative design to support the data analyzed in the first phase (Creswell, 2013). For the quantitative component, a quasi-experimental non-equivalent design will be used with experimental (feedback) and control (no feedback) groups. Creswell (2013) justifies the use of this design since the study cannot create artificial groups and random assignment is impossible. However, homogeneity of the sample’s proficiency level will be obtained, and age is controlled by the college age requirement and the prior placement test. A T-test will be applied to examine the statistical significance of written metalinguistic corrective feedback on letter writing performance. For the qualitative component, face-to-face interviews will be conducted, transcribed, and analyzed using thematic analyses to explore why Malaysian post-secondary students make errors in letter writing and to examine their perceptions of teachers’ use of metalinguistic feedback.

Ethical Considerations

The study will be conducted after receiving permission to access the research site. The department head, teachers, and target participants will be informed about the study. Participation will be voluntary, and privacy will be protected.


Two classes from Kolej Yayasan Felda (20-25 per class), ages 18-20, will participate in this research. They will have an intermediate level of English proficiency based on the college’s placement test.

Selection Criteria                                              

Convenience sampling will be done since random assignment is not possible. It would also be unethical to disrupt classes which are established prior to the study. However, students within the target population are homogenized by the Standard English proficiency placement test to be conducted by the college. The researcher will select intermediate-level students who possess basic English writing skills and can understand and correct their own errors through metalinguistic cues.

Quantitative Experimental Non-Equivalent Design

The quantitative component of this study will investigate whether metalinguistic corrective feedback affects the technical quality of job application letters written by intermediate level, L2, Malaysian post-secondary students. This component will be conducted in four stages using a standardized testing instrument. It will include: (1) a pre-test given to all participants to obtain a baseline score; (2) random assignment  of all participants to a treatment group (receiving metalinguistic feedback) or a control group (no feedback);  (3) a post-test given to all participants for a final score; and (4) a delayed post-test to measure long-term effect.

  • Testing Instrument

The standardized testing instrument will be patterned on the CEFR guidelines to measure the linguistic quality of job application letters. Specific guidelines of what to be written will also be instructed. Pilot tests will be administered to other student groups, and college teachers will provide feedback on the instrument to ensure the test’s validity and reliability.

  • Data Collection

The pre-test will be administered by the researcher to all study participants within a 60-minute time frame. Following the pre-test, two raters will review and score the test according to their respective experimental group: (1) the treatment group receives metalinguistic corrective feedback and a score, and (2) the control group receives a score but no feedback. Following the pre-test, participants will receive the scored and treated test, have one week to review it, and then be tested again with an identical or similar instrument for the post-test.

  • Test Scoring

A standardized rubric developed by City and Guilds, UK, will be used to rate the students’ writing performance. Two teachers will rate the pre-test and post-test results from each group to assess the differences within and between groups, and to ensure reliability of scores. Inter-rater reliability will be accounted for using two methods: (1) calculating the kappa statistic for pre- and post-test scores among all raters, and (2) percent agreement between pre- and post-test scores among all raters.

  • Treatment set-up/ Operationalization

The quasi-experimental study will last 5 months. This is an optimum time span to facilitate the quasi-experimental design. In the first week, the pre-test will be given to both groups. Students will be asked to write a job application letter. Written outputs will be rated by two teachers individually. After a week, papers with only scores will be returned to the control group while papers with scores plus metalinguistic corrective feedback will be given to the experimental group. Error codes such as AR= article, VA=Verb Agreement, and SP= spelling will be placed beside the sentences/phrases in which the error is found by the rates in the experimental group. A post-test will then be administered in the third week of the same procedure to assess the performance of all students. Their writing will be rated again by two teachers.

To test if there are any long-term changes, a delayed post-test will be given to the students in the sixth week with the same grading procedures.

  • Data Analysis

Scores will be entered into SPSS and a t-test will be used to determine any significant differences between the treatment and control groups and within experimental groups. The independent and paired two-tail t-test will determine whether there are significant differences between the control and treatment groups at the pre-test stage.

A t-test will then be used to compare the scores of the pre- and post-test between the treatment and control groups.

Qualitative Design

The qualitative component of this study seeks to determine how students view their teacher’s use of metalinguistic corrective feedback.

  • Interview Instrument

A semi-structured interview based from the results of the quantitative study will be designed to explore the students’ view on metalinguistic feedback received from the teacher, and whether this feedback has been helpful or not. The conversation will be free-flowing and natural in order to gain deeper insight into student responses (Patton, 2004).

  • Data Collection

Through stratified purposeful sampling, students in the treatment group who scored above average, average and below average, will be invited to participate in a 15-25 minute, semi-structured interview with the researcher. This is to capture the major variations of the students’ views (Patton, 2004). Interviews will be digitally recorded and transcribed by a qualified transcriptionist. Transcriptions will be assigned with random case numbers and the individuals’ names will be removed to protect the anonymity of the participants.

  • Data Analysis

Thematic analysis will be performed to analyze the qualitative data. This is to explore students’ understanding of their teachers’ use of metalinguistic corrective feedback other than the usual feedback they receive. It will seek to gather students’ perspectives and views on this type of feedback and its efficacy.

  • Data Analysis Procedure

Transcripts from the semi-structured interview will be uploaded into QDA Miner Lite- Provalis 2015 and coded into themes and subcodes. Subcodes will be used to interpret the deeper meanings in themes.

Coding will be done in three steps – the first coding cycle, second coding cycle, and post-coding cycle. A first coding cycle involves reading over the material for any information to become more familiar with the data and the voice of the participant and to reflect on the questions and answers. The second cycle will contain the majority of the coding – the researcher looks for deeper meanings, categories themes, and provides loose interpretations of the data. Depending on the amount and richness of the data, the researcher may review it several more times before moving to the post-coding cycle. The post-coding cycle will allow the researcher further reflection on and interpretation of the data, patterns, and form theories (Saldaña, 2013).

The researcher will use the three coders from the quantitative component to code the interview responses. Rates will consult with each other over their coding, providing opportunities for deeper analysis of the perceptions of students.

Final Interpretation (Quantitative and Qualitative)

This section will provide summary of the quantitative and qualitative results and discuss to what extent and in what ways the qualitative result helps to explain the results in the quantitative study.


Through the application of these qualitative and quantitative strategies, this study aims to contribute to the existing literature on written metalinguistic feedback in the context of Malaysian young adult learners. If undertaken, this study will be a significant contribution to the literature examining corrective feedback to those learning a second language (L2 learners). As previously mentioned, there is very little information currently available on the effects of metalinguistic corrective feedback on improving the written English of intermediate-level young adult English learners. It is hoped that this study will shed light on the appropriateness of this form of feedback for improving young adult L2 learners’ written English skills and provide a better insights on how these students could be more effective in writing application letters. Further research will be needed to confirm the results of this study, and to verify if these results can be applied to populations other than young Malay adults ages 18-20.

Share with your friends
Order Unique Answer Now

Add a Comment