Description Quantitative, qualitative, and mixed methods research

Introduction: developmental issues in pragmatics

-journalpapers.org- Over the last three decades, a number of studies in second language (L2) pragmaticsdwhether qualitative, quantitative, descriptive, or quasi-experimental have addressed developmental issues of pragmatic competence (for a review, see Taguchi & Roever, 2017). Developmental studies go beyond just the description of pragmatic language use. They intend to capture changes in pragmatic competence and explain these changes by examining inﬂuences eboth contextual and indi- vidual that may be related to the changes. Some questions addressed in this process include:

Past research that explored these issues largely falls into three areas: cross-sectional, instructional, and longitudinal. Of these three areas, cross-sectional research has generated the most ﬁndings. Cross-sectional studies collect data from two or more sections of a sample based on differences in general proﬁciency, span of formal study, or length of residence in the target community. Any differences in pragmatic performance gleaned from the comparison across groups are attributed to changes that learners exhibit at different stages of their L2 learning and thus are considered to provide developmental insights. Findings from cross-sectional studies generally point out that higher proﬁciency and length of residence are associated with pragmatic abilities, but they do not guarantee same levels of development across individuals, contexts, and pragmatic targets (Xiao, 2015a).

ilustrasi

Similar to cross-sectional studies, instructional studies are concerned with change and factors affecting the change, but more precisely, they focus on direct instruction as the factor and examine its impact on learning outcomes e the change in pragmatic knowledge from pre-to post-instruction. Some instructional studies are also considered longitudinal, including delayed post-tests and implementing long-lasting interventions (e.g., Alcon-Soler, 2015; Cunningham, 2016). Instructional studies are largely quasi-experimental, comparing intact groups of learners who received instruction with those who did not, or examining two or more groups under different treatment conditions. Existing ﬁndings unanimously point to the positive role of instruction in pragmatic development (e.g., Halenko & Jones, 2011; Li, 2012; Nguyen, 2013). Other generalizations that emerge in the literature include: (1) Explicit instruction is effective when it involves direct metapragmatic information and production practice of pragmatic forms; and (2) Implicit instruction can be effective if instructional tasks promote noticing and processing of pragmatic features (e.g., having learners respond to input or compare examples to induce rules behind them) (Taguchi, 2015).

Quantitative and qualitative approaches in longitudinal L2 pragmatics research

Quantitative studies can reveal how L2 learners' pragmatic competence changes over time through systematic, cyclical data collection. Researchers administer a comparable task to elicit performance from a group of learners at certain time intervals and quantify their performance with numerical indices (e.g., scores, ratings, and frequency counts). Then, they compare group-level performance across time points, often using inferential statistics such as repeated-measures ANOVA or mixed-model analysis. By controlling the task, data collection procedure, and context of study (e.g., same learner group in the same environment), quantitative research seeks a causality relationship between time and change. Systematic data collection, ﬁxed time intervals, statistical analyses, and effect sizes allow researchers to determine whether learners make a signiﬁcant gain over time and how large the gain size is on the given time scale. By incorporating contextual and individual variables into design, researchers can also investigate factors contributing to the gain, which in turn generates explanations for pragmatic development.

The beneﬁt of using a qualitative approach is the approach's open-ended, exploratory stance in which researchers can commit to gaining an understanding of reasons behind a phenomenon, rather than proving or disproving a hypothesis. Unlike a predictive approach using a linear analysis of pre-selected variables, qualitative research can deal with a greater number of elements that emerge in the data and use them as a whole as evidence toward interpreting observed changes

Mixed methods approach in longitudinal L2 pragmatics research

Purpose and types of mixed methods research

According to Creswell, Plano Clark, Gutmann, and Hanson (2003), mixed methods research involves “the collection or analysis of both quantitative and qualitative data in a single study in which the data are collected concurrently or sequentially, are given priority, and involve the integration of data” (p. 29). This deﬁnition underscores several characteristics of research design and data collection in mixed methods. First, quantitative and qualitative data are collected either concurrently or We may ﬁnd something in quantitative data and ﬁnd something different in qualitative analysis. Tashakkori and Teddlie argue that such divergent ﬁndings are valuable because they generate new understanding of a phenomenon.

While there is uniformity in the deﬁnition and beneﬁts of mixed methods research, there are different design types in mixed methods (Creswell et al., 2003):

Sequential explanatory: Quantitative data is collected ﬁrst, followed by qualitative data that can explain the ﬁndings from quantitative data (e.g., after assessing pragmatic competence at group-level, following up on several participants to gain understanding about their characteristics).
Sequential exploratory: Qualitative data is collected ﬁrst, followed by quantitative data to interpret qualitative ﬁndings(e.g., collecting interview data to identify recurrent themes and then using those themes to develop survey items). (3) Sequential transformative: Quantitative or qualitative data are collected in sequence for the purpose of changing an existing policy or practice.
Concurrent triangulation: Quantitative and qualitative data are collected simultaneously with equal emphasis and are used to cross-validate ﬁndings
Concurrent nested: Quantitative and qualitative data are collected simultaneously, but one type of data is more pre- dominant than the other. Less predominant data is nested (embedded) within more predominant data and used to enrich description of the data.
Concurrent transformative: Quantitative and qualitative data are collected simultaneously with equal emphasis, and are used for the purpose of changing an existing policy or practice.

Conclusion

Another future challenge relates to the construction of a concrete research design using mixed methods. I encourage researchers to ask how best they can answer the research questions they propose and think about the best possible evidence that they can collect to answer the questions. Such evidence will naturally involve quantitative and qualitative data because more evidence leads to stronger inferences. This evidence-based perspective will help researchers think how to intersect various data sources to answer research questions and how to design a study accordingly. Given the complex nature of pragmatic competence, studies need to be designed in a way that multiple data sources can facilitate a multi-level analysis of context, individuals, and pragmatic development. Quantitative and qualitative data in a triangulated format can shed light on the complexity of pragmatic development in which multiple factors are interconnected and jointly inﬂuence the develop- mental trajectories as they change.

Description Quantitative, qualitative, and mixed methods research