Constant comparison
Constant comparison is the grounded theory method of analyzing qualitative data by setting incidents, codes and categories side by side to see where they match and where they differ. Those comparisons decide when codes merge, split or grow into categories.
How it works
As an analyst codes a new passage, they compare it with every other passage carrying the same code: is this really the same thing, or something different that deserves its own code? They also compare incidents across participants, codes with categories, and categories with each other. Each comparison either confirms a category, sharpens its definition or splits it.
Barney Glaser and Anselm Strauss set out the method in 1967 as the engine of grounded theory, and it runs alongside data collection, so what the comparisons turn up shapes who is interviewed next. Outside grounded theory, the same habit keeps any codebook honest, because it stops a code from drifting as more data arrives.
A worked example
In interviews, 5 admitted students who turned down a university's offer say they "felt rushed". Setting a sixth account beside those five, the admissions researcher sees two kinds of rush, pressure from a counselor's calls and a deposit deadline printed in the offer letter. She splits the code. Of 14 later mentions, 9 concern the letter, so the fix moves from counselor training to the letter template.
Related
Automate this in Pulse
Pulse's themes gather like passages together, and the grounded theorist's work of comparing within a category, asking how this complaint differs from the last one, can start from those groups.