Academic journal article The Qualitative Report

Applying Constant Comparative Method with Multiple Investigators and Inter-Coder Reliability

Academic journal article The Qualitative Report

Applying Constant Comparative Method with Multiple Investigators and Inter-Coder Reliability

Article excerpt

When qualitative research methods are used, data analysis may be completed by an individual or a group of two or more people. Researchers accustomed to completing independent data analysis may be surprised by the large amount of additional time and effort that working with a research group requires. Collaboration adds complexity to the work of data analysis and formulating findings, making a collaborative qualitative study more labor intensive (Miles & Huberman, 1994). Additional coordination and iteration are required for the qualitative coding process for creating themes, analyzing for meaning, and drawing conclusions. When members of a research team are geographically separated and working in a virtual environment, data analysis may be more challenging. However, the collaboration provides several benefits that derive from the additional perspectives provided by multiple researchers. In striving for consensus in the findings, the nuances in meaning brought by multiple researchers adds richness to the analysis by prompting deeper analysis. Inter-coder reliability (ICR) can be used to drive towards consensus but was found to be more suited for identifying nuance and significant meanings in the qualitative data. This paper explores the experiences of two geographically separated researchers who applied Constant Comparative Method (CCM), based on grounded theory. The researchers applied action research to formulate a deliberate 10-step method for coding data, creating meaning, and structuring an exploratory model that represents findings. Collaboration was facilitated through synchronous online video discussions and email exchanges to work through analysis activities between the two researchers.

Literature Review

Literature on qualitative research, and specifically on the CCM methodology used by the researchers performing this study, reveals a diversity of positions that reflect the richness of qualitative research (Strauss & Corbin, 1998). There are supporters and opponents to qualitative research in general and CCM in particular. This review begins with a basic explanation of the approach that differentiates qualitative research from quantitative; then explores the methods used in qualitative research to address issues common to quantitative researchers involving validity and reliability. Finally, the review will focus on the literature concerning advantages, disadvantages, and potential roles of ICR measures in CCM.

Inductive Approach

The original purpose of qualitative methods was to design a structured approach for generating new theory that purports to explain an experience or phenomenon for which current understanding is inadequate. Qualitative research uses inductive reasoning (i.e., developing explanations from information) rather than the deductive (i.e., using theory to predict outcomes based on information) to draw conclusions from data. It explores a deliberately selected set of data, such as interviews, observations, or video/audio logs, to identify patterns that can be linked causally in a model or theory (Thomas, 2003). Models generated by qualitative theory can be tested using quantitative methods to provide further support for the theory. Quantitative research uses existing theory to generate a question or hypothesis that can be tested empirically (Curry, Nembhard, & Bradley, 2009).

Grounded Theory

Grounded theory is a qualitative research method developed to facilitate discovering patterns in data (Glaser & Strauss, 1967). It uses a systematic approach to review participant views collected from an experience in order to allow patterns and themes to emerge over multiple passes through the data. Strauss (1987) further elaborated on the data analysis methodology, creating CCM, in which the researcher developed codes while reviewing transcripts or other verbatim data to identify constructs, and iteratively compared texts identified with the same code to ensure they were representative of the same construct. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.