The constant comparative method (CCM), also sometimes referred to as the constant comparison method, is a four-step method for qualitative theory building research. The constant comparative method is part of the foundation of Grounded Theory. However, while the inception of this method is often incorrectly conflated with the discovery of grounded theory by Glaser and Strauss in 1967, it actually predates grounded theory. The core tenet of the constant comparative method is easily summarized by a quote from Glaser (1965): “While coding an incident for a category, compare it with the previous incidents coded in the same category”.
This leads to a frequent back and forth between the different data collected until this point, and it also is strongly coupled to the concept of theoretical sampling, where the sampling is driven by the comparisons made between the incidents of each theoretical construct and the thereby emerging themes or contradictions.
The process of Constant Comparative Method
The constant comparative method, according to Glaser (1965), consists of the following four process steps:
Comparing incidents: Analyze and code each incident in the data across multiple categories, comparing new incidents with previously coded ones within the same category.
Integrating categories and properties: As coding progresses, integrate categories and their properties through continuous comparisons, refining understanding of relationships between incidents and categories.
Delimiting the theory: Solidify the theory by reducing and clarifying categories and properties, focusing on relevant aspects and eliminating non-essential details.
Writing the theory: Compile and articulate the developed theory, ensuring it reflects the integrated categories and properties derived from the analysis.
Levels of comparison in the constant comparative method
The types of comparisons that are necessary avary from the type of study you set out to do, and the research question you want to answer. While the process described by Glaser, and outlined in the previous section is very high-level, Boeije (2002)provides a more detailed example on the analysis process broken down into different levels of comparison. Boeije (2002)explicitly acknowledges that the specific number of steps, or types of comparisons, is dependent on the data, but we recommend giving this example a read and then define your own dimensions of comparison if you want to cite the constant comparison method as your analysis method of choice. You should think about the following dimensions when defining your own levels of comparison:
Within interview: Comparsisons between parts of the same interview are usually done within the open coding phase of the analysis and help to uncover instances where an interviewee contradicts themselves and to make sure you have a good understanding of the individual perspective and narrative told by the individual participant. Comparing how different parts of the interview present aspects of a particular theoretical construct may also lead you to split up a code into two more precise ones or add a new dimension to the analysis as well as prepare follow-up interview outlines. The added or modified codes are the result of these types of comparisons.
Within subgroups: Sometimes your particiapnts can be clearly separated into subgroups along one ore many dimensions and it is important to compare new instances of a code with other instances within this category with an eye on consistency or a group narrative. Examples of such a dimension could be a single case within a multiple-case case study. Or multiple participants with a working relationship or a family relationship. These comparisons also help structure your code system and are part of your axial coding.
Across subgroups: Comparing instances of a code from one subgroup to those found within another is important to identify typical manifestations of it as well as variants, descrapancies and contradictions. These feed into your theoretical sampling and further drive the building of categories. The level of analysis here is still between individual participants and the results are documented in memos as well as changes in the code system.
Between subgroups: Taking the analysis to a new abstraction level it is also important to compare groups of paticipants against each other. This type of comparison can also be multi-dimensional, for example by considering multiple types of categorization such as belonging to a certain case in your case study and having a particular marital status and having experienced a particular aspect of the phenomenon compared to not having experienced this aspect of the phenomenon.
The results in the analysis artifacts are usually also documented with memo writing and in changes to your code system.
Conclusion on the Constant Comparative Method
Tesch (2013)describes comparison as the main intellectual tool in qualitative analysis, writing “the method of comparing and contrasting is used for practically all intellectual tasks during analysis: forming categories, establishing the boundaries of the categories, assigning the segments to categories, summarizing the content of each category, finding negative evidence, etc. The goal is to discern conceptual similarities, to refine the discriminative power of categories, and to discover patterns.” While the constant comparison method is frequently cited, the precise steps taken and the comparisons made during analysis are often only described in a vague way. We recommend you document the different types of comparisons you did during your analysis in your audit trail and also lay them out transparently in your writing.