Verbatim Responses

A verbatim response refers to what an interviewer records as an answer to an open-ended question when writing down the exact words spoken by the respondent. Open-ended questions are those that do not provide a respondent with predetermined response choices and instead allow, expect, and encourage a respondent to answer in his or her own words.

Sometimes open-ended questions have “precoded” response categories for interviewers to use, but these are not read to a respondent. In these instances interviewers are trained to listen to the answer a respondent gives and then “fit” it into one of the precoded choices. However, this is not always possible to do for all given responses, so even with precoded open-ended questions, some verbatim responses that do not fit into any one of the precoded categories must be written down by the interviewer. Most open-ended questions do not have precoded response choices for interviewers to code; thus in most cases with an open-ended question, the interviewer must write down what the respondent says.

In some cases a researcher may allow interviewers to summarize the gist of what the respondent says in response to an open-ended question. However, in other cases it is important that the exact words spoken by the respondent are recorded by the interviewer. It is in these cases that interviewers are trained and expected to write down the complete verbatim response. Interviewers are trained to “slow down” respondents who are speaking too fast in answering an open-ended question for which a verbatim response must be recorded by the interviewer. Interviewers also are trained to use probing techniques to get more detailed replies to open-end questions.

After data collection has ended, the verbatim responses that interviewers have recorded often are coded (similar to content analysis) so that they then can be analyzed via quantitative statistical techniques. This is a labor-intensive and expensive process, if done reliably. Verbatim responses also are used in survey reports to help illustrate the statistical findings gleaned from coded open-ended variables by putting a “human face” to them. This is done without identifying the respondents whose verbatim responses are used, so as not to violate the confidentiality pledge given to the respondents.