The process by which we identify and denote a numeral to the responses that have been given by the respondent is called the coding. The main purpose behind doing the coding is that it helps to facilitate the researchers use so that the answers can be interpreted, classified and recorded from the questionnaire onto the excel sheet in the computer. What is essential and required is to give a numeric code to the data that is categorical in nature. Even the questions that fall into the open ended category and are in statement form should be attempted to be categorized into numbers. The main reason for doing so is that the data can be quantified and represented in the forms of charts and graphs.
The method that is used for coding of the structured questions is comparatively easier because the category of responses have been decided in advance. All that the researcher has to do is to assign a specific code for each possible answer and assign the fields and columns for which the various response codes have to be noted. There are different kinds of coding methods that have to be followed for different kinds of questions. They are:
Dichotomous Questions: For questions that fall in the Dichotomous category, that are of the nominal category we can give binary responses in 0 and 1.
Ranking questions: For the different types of ranking questions, where a lot of objects have to be ranked, the researcher will have to make a lot of columns and the column numbers have to equal to the number of objects that have to be ranked.
Check list/multiple responses: For those kind of questions for which multiple responses can be there should be a separate column for each type of response.
Scaled Questions: There are sometimes questions that are of scaled category. They usually are of interval scale. The coding in this case would have to be done for the different kinds of responses that can be there. Though for comparative scales, the coding instructions would be slightly different.