Friday, October 30, 2009

comps reading - What Good is Polarizing Research into Qualitative and Quantitative?

The purpose of this paper co-written by Kadriye Ercikan and Wolff-Michael Roth is twofold:
  1. to demonstrate that this polarization is not meaningful or productive for education resarch
  2. to propose an integrated approach to education research inquiry
To achieve such integration, the authors argued that the research questions asked should determine the modes of inquiry that are used to answer them.

First the authors believed that polarization in research is confusing to many people and tends to limit research inquiry, often resulting in incomplete answers to research questions and potentially inappropriate inferences based on findings. To demonstrate such polarization, the authors unfolded from four aspects:
  1. The problem
  2. Existence of qualitative and quantitative characteristics in phenomena: as data are representations of phenomena in nature, society, education, and cultures, the main determining factor of the classification of research activities into qualitative and quantitative is the nature of the data.
  3. Objectivity and subjectivity in constructing data: both types of research activities (qualitative and quantitative) involve subjective judgments. Basically, the subjectivity involved in data construction phase depends on "who is involved in scoring," "their level of experiencing in scoring," "the scoring rules," and "the students response or products." Any data that are constructed arise through an interpretation model that involves subjective judgments. The processes involved in constructing data may be quantitative or qualitative in nature and include three dimensions: data sources, interpretation model, and data.
  4. Generalizability
Based on the following three reasons, the authors concluded that polarization in research is problematic.
  1. All phenomena and all knowledge simultaneously have quantitative and qualitative dimensions
  2. The distinction between objectivity and subjectivity, normally associated with that between quantitative and qualitative research, is neither accurate nor useful
  3. Generalizability is not a feature of mathematization but a description for the tendency of inferences to go beyond the context and participants involved in the research.
To counter these problems, the authors proposed moving beyond polarization by:
  1. using a different classificatory continuum based on the relational terms "low inference" and "high inference"
  2. emphasizing a focus on the research questions
  3. encouraging the collaboration of researchers with expertise in forms of research formerly labeled quantitative and qualitative

No comments: