It appears qualitative content analysis has come a long way. Formerly attributed only to quantitative research, qualitative content analysis is gaining respect because of its ability to address some of the weaknesses of statistical analysis and probabilities. For instance, unlike quantitative statistics, qualitative data analysis, qualitative study allows for the subjective interpretation of data using systematic coding classifications to better identify themes or patterns (Hsieh & Shannon, 2005, p. 1278) and analyze data within the context of communications (Mayring, 2000, p. 2).
Question: How does qualitative data collection and analysis methods differ from the quantitative methods?
Qualitative Content Analysis and Why
Zhang and Wildermuth (2009) explain qualitative content analysis is better suited to allow researchers to condense large amounts of data into categories, draw inferences and interpretations, then carefully examine information gathered and apply the process of either inductive or deductive reasoning repeatedly until the data makes sense. Further, Zhang and Wildermuth identify the methods used to code and categorize qualitative data and that allows researchers to ground their examination of a topic of inquiry using flexible or standardized categories in their attempts to generate valid inferences and interpretative theories regarding the observed phenomenon (p. 1-2).
Best Practices for Coding Qualitative Data
Zhang and Wildermuth supported their position by explaining how researchers could condense raw data into categories or themes to facilitate researcher examination and comparisons (p.2). These researchers advocate qualitative researchers use eight key steps to prepare, define, analyze, test and retest then assess raw data. Zhang and Wildermuth recommend the below key steps for qualitative coding data:
Step 1. Prepare the data. Convert the various types of data collected into text for examination.
Step 2. Classify data. Convert text into “chunks” or units of information based on content or context.
Step 3. Coding Scheme. Chunks of information are organized into categories using context and content then coded according to a scheme.
Step 4. Test Coding Scheme. A sample of the coded data is used to validate the levels of consistency and sufficiency. The process is iterative until all doubts about the accuracy of coding is satisfied and create a manual.
Step 5. Code Data. Sample coding plied to all data collected using the coding manual as a guide.
Step 6. Assess Coding Accuracy. Coding schemes are rechecked for consistency to ensure all mistakes and problems have been addressed or mitigated.
Step 7. Draw Conclusions. Themes and patterns allow exploration of properties to make sense of the data using at least three different approaches (triangulation) and by examining relationships between categories, uncovering patterns, and testing categories.
Step 8. Report Methods and Findings. Inferences are drawn from data, implications, limitations, and reported with recommendations (Zhang & Wildermuth, 2009, p. 2-5).
But …is it right?
Zhang and Wildermuth warn that researchers should ensure they remain transparent throughout the above process by taking thorough field notes and recording their data collection, coding, analyses, and reporting methods (creditability). Researchers are also advised to generate a hypothesis that can be applied in another context or different settings (transferability). Finally, Zhang and Wildermuth recommend qualitative researchers ensure their internal processes account for changing conditions (dependability) and that their method can be confirmed by others who read or review their research results (confirmability) (p. 6)
Ravitch and Carl also noted the importance of creditability, transferability, dependability, and confirmability for qualitative researchers. These authors indicated that qualitative researchers are held to a different standard in that qualitative researchers must meet the standard of trustworthiness because of the subjectivity of their interpretations of data. By comparison, quantitative researchers seek validity, reliability, and objectivity for their understanding of statistical probabilities. The former is much more prone to bias and vulnerable to positionality .
Creditability Through Triangulation
Therefore, Ravitch and Carl ensure triangulation processes are embedded in their data collection and analyses processes such as:
- Using different strategies and methods during data evaluation;
- Searching for and collecting data from as many data sources as possible using different sampling strategies and individuals at varying times and places;
- Involving multiple researcher perspectives during interpretation of data, identification of themes and patterns, and reporting findings and conclusions;
- Framing the theoretical underpinnings of the study and noting how it compares or contrasts with previous works and why; and
- Intentionally and systematically including a wide range of participants and diverse groups (Ravitch & Carl, 2015, p. 186-200).
While exploring the above advice from Zhang and Wildermuth and Ravitch and Carl, I assumed qualitative coding of participant interviews would be more than a notion. I was right. My first interview field notes coding experience was quite a challenge. I struggled through it, however, and preserved until I reached an elementary level of coding accuracy sufficient to get a passing grade on the assignment.
Lessons learned? Garbage in. Garbage out. In other words, qualitative data collection and coding takes practice. And, I discovered I definitely need more of that.
Hsieh, H.-F., & Shannon, S.E. (2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15(9), 1277-1288
Mayring, P. (2000). Qualitative content analysis. Forum: Qualitative Social Research, 1(2).
Ravitch, S. M., & Carl, N. M. (2015). Qualitative research: Bridging the conceptual, theoretical, and methodological. Sage Publications.
Zhang, Y., & Wildemuth, B. M. (2009). Qualitative analysis of content. Applications of social research methods to questions in information and library science, 308, 319.