Before any important study that will rely on telephone interviewing as a data collection tool, it is important that the questionnaire has been thoroughly tested. I was always surprised when clients turned down the opportunity to sit in on a telephone interviewing test, because the testing procedure provides opportunities to refine the questionnaire so that valid and meaningful information will be obtained from the interviews.

Testing a telephone interview questionnaire can be a tedious task, but the time spent during the testing stage can be critical to the success of the project. One should feel comfortable about the validity of the data only if they are certain that the questionnaire was properly written and administered.

A proper test requires patience and concentration, and it is important not to lose sight of the purpose of the occasion. Boredom and frustration can occur while waiting for interviewers to find qualified respondents, so that all portions of the questionnaires can be tested properly. In addition, trying to decide on the changes in the questionnaire and communicate them to the interviewers can fray one's nerves.

The first task is to properly train the interviewers who will test the instrument (survey). Normally a telephone interviewing facility has a number of personnel who have experience with this type of monitoring. Some clients might question the use of special interviewers for this type of work, but I always preferred handpicking these people. At this stage of the project the purpose is not to test overall interviewing skill and knowledge; that can be done during the actual interviewing procedure. What we all want is the best questionnaire possible, developed in the shortest period of time. Experienced interviewers are the ones most likely to accomplish this.

The training should be similar to that which the full interviewing staff will undergo when the final questionnaire is ready. Interviewers should be allowed time to ask questions, and dummy interviews should be done so that the personnel feel comfortable with all elements of the questionnaire.

Interviews vary greatly in degree of difficulty, therefore it is important to listen carefully not only to the respondent but also to the interviewer. If the interviewer is having trouble following skip patterns, pronouncing words, or with other parts of the questionnaire, you should make a note to discuss it in the debriefing.

One word of caution: The interviewing test period is not the time to evaluate the research firm you have hired to conduct the project. That decision has already been made and you want to focus on the interview itself. Evaluating the interviewing staff can be done during the regular interviewing process but right now keep your focus on the questionnaire.

I preferred to work with only two or three interviewers at a time on a test. I would record the answers on a questionnaire and meet with the interviewer immediately after a completion for a review. First we compared the completed forms to see if there were any differences. Then we discussed any apparent problems which had occurred during the interviewing process. If it appeared changes should be made in the questionnaire, we did so immediately so that the revision could be included in the next interview. A word processor came in handy at this time because we frequently used paper copies in our tests.

Occasionally an interviewer will complete an interview which has not been monitored. I like to review these questionnaires immediately also, since experienced interviewers can be a great help because they have usually developed a sensitivity to the people they are querying. In particular I have found they have a keen awareness of one of the greatest problems in telephone interviewing: the overlong interview. They know when the respondent has grown weary and is about to either terminate the interview by hanging up or, even worse, begin to provide inaccurate information in order to hurry the interview along.

After the interview test is completed, a debriefing session is held with the interviewers. I have each interviewer review the survey, question by question, so that nothing is missed. Discussion of the questionnaire among the interviewers is encouraged and although their statements and opinions are not final, they carry a lot of weight in reviewing the questionnaire.

At this point the testing process is almost completed. The next draft of the questionnaire is tested, using the same procedure. Usually if the initial test was done correctly, few revisions are necessary before beginning the final interviewing process.

Testing a telephone questionnaire is time-consuming, but having completed the process I always felt more confident that the results obtained would be valid and useful.