Coding is the process of categorizing qualitative data so that the data becomes quantifiable (i.e., measurable). Of course, before a researcher can code raw, qualitative data (such as a taped interview, newspaper archives, weblogs, or ethnographic narratives), he or she must pose a research question. How data is coded depends entirely on what the researcher hopes to discover in the data; the same qualitative data can be coded in many different ways, calling attention to different aspects of the data.

The process of coding is one of developing categories and sub-categories, as the above graphic attempts to illustrate visually. For example, if I am researching students at an adult education center in the inner-city, I might ask the following research question: What outside exigencies influence a student to graduate and what influences him/her to drop out? I then interview several classrooms of students, which provides me with the raw data. Coding the data, I decide how to pair different aspects of the students' responses to different categories: Family issues (F), Work responsibilities (W), Lack of transportation (T), and Negative attitude (N) might be several reasons I discover for why adult students drop out. These are my first categories. Then, I return to the data and, if necessary, refine the categories. Under Family issues (F), I might discover that this includes Lack of childcare (F-LCC), Lack of spousal support (F-LSS), and Violent home life (F-VHL). If my data requires it, I could even break up these sub-categories into deeper and deeper categories, refining the data so that it better answers my research question: what influences adult learners to graduate or drop out of adult school? With the categories set, all that remains is simple math and statistics.
Categories are ad-hoc, so it is important to be flexible with them, allowing the data to take you where it wants. Somtimes it becomes necessary to break up the data into separate chunks (i.e., separating student interviews into "daytime" and "evening" interviews, thus creating your first category by breaking up the data). Coding data is thus an art as much as it is a science.

Given the huge number of fields (pychology, education, sociology, to name only a few) that demand the coding of qualtiative data, it is no surprise that many software companies have developed software to assist researchers with their coding: QSR International, ResearchWare, Atlas.ti, and Provalis Research are just some of the companies that offer researchers powerful software for qualitative coding, data mining, and content analysis.

The following links will take you to more detailed information on the many different coding processes a researcher might employ.
Analyzing Qualitative Data, compiled by the University of Wisconsin Extension Program
Coding Qualitative Data, a step-by-step process by Sonja K. Foss and Wiliam Waters
Learning Qualitative Data Analysis, a web-based tutorial by the University of Huddersfield, UK
Qualitative Data Analysis, a detailed chapter posted by Burke Johnson at the University of Southern Alabama