Mixed Method Research Design - Language Research

Mixed Method research design in language research

1. Mixed Method Research Design Joanna M. Noveno

2. Mixed Method Explained Mixed methods research—also known as multi-method, combined methods, mixed research, or triangulation (Mackey & Gass , 2015)—is a strategy of inquiry that allows the researcher to explore a research question from multiple angles potentially avoiding the limitations inherent in using one approach, quantitative or qualitative, independently.

3. Mixed Method Explained Mixed-methods research (MMR) is a new research approach in which both quantitative and qualitative data and analyses are used in a single study to address more complex research issues. In applied linguistics research, the most common way mixed methods are described is “triangulation” of multiple sources or methods (Mackey & Gass , 2015).

4. Types of Mixed Methods Studies Concurrent and Sequential Designs

5. Types of Mixed Methods Studies Concurrent Design the researcher collects both quantitative and qualitative data at the same time and uses the data collectively as a means to interpret the findings of the investigation (Creswell, 2003). Sequential Design involve the collection of data in multiple phases or in a set sequence.

6. Symbol System by Johnson and Christensen (2014) A plus sign (+) represents a concurrent collection of data. An arrow (  →) represents a sequential collection of data. ‘QUAL’ or ‘qual’ stand for qualitative research. ‘QUAN’ or ‘ quan ’ stand for quantitative research. Capital letters denote priority or increased weight. Lowercase letters denote lower priority or weight.

7. For example: QUAL  → quan

8. Types of Mixed Methods Studies Concurrent Designs Triangulation Concurrent Embedded Sequential Designs Explanatory Exploratory Sequential Embedded

9. Concurrent Designs Triangulation and Concurrent Embedded

10. Triangulation Designs All the data are collected in one phase and receive equal weight during analysis, allowing the researcher to “compare and contrast quantitative statistical results with qualitative findings or validate or explain quantitative results with qualitative data” (Creswell & Plano Clark, 2011, p. 62).

11. Triangulation Design

12. Lee, E. (2016). Reducing international graduate students’ language anxiety through oral pronunciation corrections. System, 56(1), 78–95. Research Background This mixed methodology study investigates the link between the type of oral corrective feedback used and language anxiety level. Research Method Both quantitative data obtained from a pre- and post-survey and classroom observations as well as qualitative data from follow-up interviews and open-ended survey questions were collected.

13. Concurrent Embedded Designs one type of data are embedded, generally meaning that it plays a secondary role to the other data, rather than both being weighed equally. In most research of this type, the focus of the design is quantitative with qualitative data collection added to support or better interpret the quantitative findings (Creswell & Plano Clark, 2011).

14. Concurrent Embedded Design

15. Sequential Designs Explanatory, Exploratory, and Sequential Embedded

16. Explanatory Design Focus on quantitative data but use qualitative follow-up data to explain quantitative results. Researchers might use an explanatory design to better understand quantitative results, especially if those results were not expected.

17. Explanatory Design

18. Example: If the outcomes of a task-based intervention showed no changes in the fluency or complexity of learners’ speech in the target language, a stimulated recall interview conducted after the intervention might elucidate some of the reasons for why no changes occurred.

19. Exploratory Design Qualitative data are collected first in order to guide subsequent collection of quantitative data. Researchers might use an exploratory design to investigate an under-explored phenomenon and define and describe variables before embarking on a quantitative design.

20. Exploratory Design

21. Mroz , A. (2015). The development of second language critical thinking in a virtual language learning environment: A process-oriented mixed-method study. CALICO Journal, 32(3), 528–553. Research Background The researcher in this study utilized a mixed methods design to investigate the emergence of L2 strategies in collaborative discourse produced by university French learners interacting in a computer-mediated setting. Research Method Qualitative data included bi-weekly observations of French classes, a focus group with five case-study students, and retrospective perception data. Quantitative data were elicited from computer-generated logs of students’ interactions in an online Second Life community.

22. Sequential Embedded Design Quantitative and qualitative data are collected in multiple phases to both inform and provide insights into potential explanations for one another. Qualitative data collected before an intervention can aid in participant selection, instrument verification, or to shape the subsequent intervention. Qualitative data collected after an intervention can help to illuminate patterns in quantitative intervention data.

23. Sequential Embedded Design

24. Préfontaine , Y., & Kormos , J. (2015). The relationship between task difficulty and second language fluency in French: A mixed methods approach. Modern Language Journal, 99(1), 96–112. Research Background The authors of this study utilized a mixed methods approach to shed light on the relationship between task difficulty and fluency. Research Method Learners participated in three task conditions that varied in demands posed on their creativity, content knowledge, and linguistic resources. Task difficulty was measured both quantitatively with questionnaire data as well as qualitatively with retrospective interview data. Fluency was measured quantitatively using a variety of fluency measures analyzed in Praat .

25. Pros and Cons to Mixed Methods Designs

26. Pros: Comprehensive Understanding: Mixed-method approaches allow researchers to gain a more comprehensive understanding of linguistic phenomena by combining the strengths of both quantitative and qualitative methods. This can lead to richer insights that might not be achievable through either method alone.

27. Pros: 2. Triangulation: By using multiple methods, researchers can cross-validate findings, enhancing the reliability and validity of their results. Triangulation helps mitigate the limitations of individual methods and provides a more robust basis for drawing conclusions.

28. Pros: 3. Flexibility: Mixed-method designs offer flexibility in research design, allowing researchers to tailor their approach to the specific research question and context. This flexibility enables them to address complex linguistic phenomena from multiple angles, adapting as needed throughout the research process.

29. Cons: Complexity : Mixed-method designs can be more complex to implement and analyze compared to single-method approaches. Integrating different methods requires careful planning, and analyzing data from multiple sources can be challenging, particularly when trying to reconcile divergent findings.

30. Cons: 2. Resource Intensive: Conducting mixed-method research may require more time, resources, and expertise compared to using a single method. Researchers may need to invest in training for both qualitative and quantitative methods, as well as specialized software for data analysis.

31. Cons: 3. Integration Challenges: Integrating qualitative and quantitative data can be difficult, particularly if the data are collected and analyzed separately. Ensuring coherence and consistency across methods requires careful attention to research design, data collection instruments, and analytical techniques.