Data coding is the process of organizing, labeling, and sorting the data into appropriate headings to make it convenient for data analysis in both quantitative and qualitative research. In data coding, responses are usually assigned numbers and symbols to make them measurable and recordable. Categories and character symbols for responses must be established before tabulation. The numbers and symbols for categorizing the data are called codes. Codes allow the researcher to reduce large quantities of information into a form that can be easily processed and analyzed, particularly by computer based statistical tools.
Coding starts with specifying the different categories or classes into which the responses are to be classified and ends with allocating codes to individual answers. The process of creating codes can be both preset and open. Some codes are predefined by the researcher. It is beneficial to start with a list of preset codes derived from the conceptual framework, the list of research questions, the problem areas, and other significant identifiers. Prior knowledge of the subject matter helps the researcher create these codes. While it is good to begin data collection and coding with preset codes, another set of codes will emerge or can be developed from reading and analyzing the data. These “emergent codes” are those ideas, concepts, actions, relationships, and meanings that appear in the data and that are different than the preset codes. The researcher tries to integrate predefined categories in the data collection forms using closed questions so that it becomes easy to transfer the data into a tabulated format later. The use of precoded closed questions reduces the cost of coding and also improves accuracy. Sometimes this is not feasible, and the researcher needs to ask open-ended questions in the survey. Including open-ended questions in a survey garners qualitative data, as opposed to the quantitative data that results from multiple-choice questions. The coding for the open-ended questions is performed after collection. Once coded, these answers can be analyzed in the same way that multiple-choice, closed questions are analyzed.
Coding is an evolving process when it comes to qualitative studies. As data is coded, the coding categories can be refined further. Researchers may add, collapse, expand, and revise the coding categories. In some cases the codes may be too broad, and subcategories may need to be added. In other cases the codes designed may be too specific and capture all possible responses. In these cases, the researcher must consider how to combine categories into a broader idea. When coding, the researcher should always make the codes fit the data, rather than trying to make the data fit the codes.
Please visit www.smstudy.com for more information.