Aging in Place Research Study - Interviewer Training

I. What makes a good interviewer?

A good interviewer is:

How do you become a good interviewer?

One word: PRACTICE. This is really the only way to become familiar with questionnaire items, and to figure out what type of approach and style works best for you. These are often highly individual matters, which take some time to discover. However, it is important that whenever possible, this takes place BEFORE actually conducting telephone interviews.

Rapport: What is it and how do we achieve it?

Rapport is the positive relationship that develops between the interviewer and the respondent during the course of the interview. One of the most important things you want to strive to do is to develop a comfortable, trusting, and relaxed rapport with the respondent. Respondents work hardest to answer most accurately and most honestly when they feel they can trust their interviewer. When good rapport is present, respondents are more careful about their responses, give us more accurate data, and enjoy the interview experience more. This all leads to better quality of the results we can report and interpret. Therefore, good rapport is key to any successful interview.

You can achieve good rapport through your verbal and nonverbal behavior. Listen carefully to the respondent, be genuinely interested in what he or she is telling you, and be sure that your body language and approach to the interview is friendly and relaxed, yet professional. The way you sit and the expression on your face can be 'heard' by the respondent if not seen. "Getting into this groove" requires practice and some trial and error.

General guidelines to help you establish good rapport

Things you need to keep in mind when doing the interviews:

  1. Sound the part
    • Convey your appearance over the phone.
    • Smile! Respondents can always 'hear' a smile.
    • Sit up straight in your chair and your voice will convey more confidence.
    • Know your project well.
    • Memorize your scripts and rebuttals so that you can have an informed conversation with the respondent rather than reading from a screen - respondents can tell the difference!
    • Personalize your introduction and give it that human touch!
  2. Act the part
    • Speak as clearly as you can and as slowly as is reasonable. Don't rush. Speak loudly enough so that the respondent can hear you but as softly as is possible.
    • Be professional - remember you are representing OSU
    • Address people appropriately, always speak politely, including please and thank you.
    • Be assertive and confident, but don't appear overly anxious or desperate to get the survey.
    • Be a good listener so that you can develop rapport with the respondent and so that you can address any concerns or hesitations that they may have
    • Use the word interview, rather than survey, as much as possible. The word 'interview' has a more professional and personal connotation than the word 'survey', even though the two words are interchangeable.
  3. Be attentive to details
    • Follow directions exactly
    • Read questions exactly as written
    • Do not put words in respondents' mouths or guess what they're saying. Record only exactly what they say.
  4. Do not influence the way in which a respondent answers - just make sure they do answer.
    • Never agree or disagree with the respondent or express sympathy for their frustration or in any other way express your views of the legitimacy or acceptability of their answers.
    • Instead of responding to the respondent with positive or negative responses like yes, no, or okay; use a neutral response such as 'I see' or 'I understand' when it is necessary to respond.
    • Be objective and calm and non-expressive. Your job is to measure their experiences and responses as accurately as possible without influencing them in any way.
    • In general, keep in mind that you influence how much effort the respondent devotes to answering carefully by the behavior you exhibit. If you are careful, attentive to detail, professional, committed and obviously interested in accuracy, this will be contagious and your respondents will be inspired to be careful and effortful, too. But if you seem not on top of what you're doing or apologizing to them for making mistakes or rushing through and not being careful, this will also be contagious, and respondents will realize that they don't need to devote much effort to generating accurate answers - they can just answer in a sloppy fashion, and that will be ok with you.
  5. Be sure that no one except the sampled respondent answers questions on the survey.
    • If a person answers the phone and agrees to do the survey, however, they are not the head of the household or the head of the household's spouse, thank them for their willingness to participate and explain that in order to ensure that this survey is a scientific, random sample, you must speak to the head of the household or the head of the household's spouse or partner. If that person is not home or not available, probe for a better time to call back.
    • If you notice at the beginning of the interview that the respondent seems to be having difficulty answering some of the questions and they want to pass the phone to a spouse or domestic partner for further clarification, reassure the respondent that it's important to you that they are comfortable answering the questions and if they would prefer to have their spouse or domestic partner answer the questions, then that is perfectly okay.

    II. Getting the interview and completing it

    An interview will not be very helpful to our study unless you gain the cooperation of the people you are assigned to interview and you complete the interviews you start. This involves seven key steps.

    1. Identify the correct respondent

      2. Your instructions will, in most cases, specify what number to call and which respondent to ask for based on household status.

    2. Approach the respondent

      3. The way you approach a potential respondent is crucial for gaining their cooperation as well as for how the rest of the interview will go. Give them enough information in the introduction to get their attention but not so much that they stop listening and get confused about why you are calling. Don't get shaken up by questions - questions are a good thing because they engage the respondent in a listening/response activity with you and it gives you an opportunity to tell them more great things about this survey.

    3. Make the request

      4. With time, this should become the easiest part of the routine, although it might not seem that way at first. You should have your introduction script memorized; however, be sure to speak as though you are simply making a polite, reasonable request, not as if you are reading a script. Remember, you want to engage the participant in a dialogue or conversation in order to make them feel involved and invested.

    4. If the respondent refuses, do not give up immediately

      5. People that are initially reluctant can often be "converted" by reassuring them about the value of the study, the shortness of the interview, and the protection of their confidentiality, and they will eventually agree to participate. If the respondent initially seems hesitant, stress the benefits he or she has to gain from the study (policy reforms to programs such as social security for example) and that they have been specifically selected to represent their peers. If these suggestions don't work, you may also try simply asking "Is there any particular reason why you are not interested or are their any questions about the survey that I can answer for you?"

      For example, if a respondent says, "Gee, I'm really in a hurry today", you might respond with "I understand that you are in a hurry and I know that your time is valuable. We are extremely interested in how you feel about economic issues." Or you could say, "I know you are, but you were specifically chosen to participate in this survey today and you represent lots of other people who are like you. We need to make sure that we get opinions from a wide variety of people, and you're important to make sure we have valid results." Also, you might try, "Yes, I can tell you're in a hurry, but we really need to know what you think about some economic issues and this is your chance to make your voice heard to policy makers." Finally, if the respondent makes it very clear that they definitely cannot do the survey at the time that you are calling, then let him/her know that you understand and that his/her participation is very important to you and that you are willing to set up an appointment to conduct the interview at a time that is most convenient for them.

      If the respondent still refuses, you must respect their right to do so and move on to the next person on your list. Don't be discouraged by these occurrences. The person you approached might have simply had too many other things on their mind or genuinely been too busy to take the survey today. Don't blame yourself for these refusals, and don't let it affect the next attempt you make.

    5. If respondent agrees, continue interview

      6. Throughout the interview, maintain a professional, yet friendly manner. In addition, you should control the interview. Do not let the respondent get too far off topic. Some respondents like to talk and will stray from answering the questions if you let them. If they do stray, acknowledge their comment with something like "I see" or "I understand" ... Well?" and then read the question you want them to answer. Be alert and attentive throughout the interview. They should feel like their opinions are important and worth listening to (because they are!).

    6. If respondent needs a probe

      7. Sometimes a respondent has difficulty with a question or answers it inappropriately. Then it's time to use an appropriate probe. 'Probes' are questions or techniques used to get a single answer to the question you're asking, even if the participant hesitates or seems unsure. They are especially necessary when a respondent gives a "don't know" response. Some situations and their probes are elucidated below. You will learn to develop your own probes in response to the situation.

      I Don't Know: If a respondent says "Gee, I don't know", just pause and wait silently and patiently - "I don't know" sometimes means the person is playing for time while thinking about the question, not that the person truly has no answer to give. Pause for what will feel to you like a REALLY LONG time, even though it will only be 5 seconds. Count "one one-thousand, two one-thousand, three one-thousand ..." slowly in your head to time the five seconds. That usually inspires a respondent to give an answer. If you pause for that long and the respondent hasn't answered, then you can say something like "I'd be grateful to get the best answer you can give me about this -- shall I re-read the question one more time?" Because this survey deals with finance, there are many questions that ask for specific dollar amounts. Respondents will be quick to say that they "don't know" the exact amounts. It will be your job to reassure the respondent that "the best guess or best estimate will be just fine". If the respondent still insists that he or she cannot make a guess or estimate, simply mark "don't know" for that question and continue with the survey.

      More than one answer: If a question, for example, requires a specific number entry, and the respondent gives an answer such as "between 20 or 30" then you would want to probe with something like "I see, would that be closer to 20 or closer to 30?" You never want to pick an answer for a respondent -- get them to pick a valid answer.

      Answer not on the scale: For questions with a scale answer, it is important that the respondent give an answer that matches the scale choice exactly. For example, if the valid answers are 'Excellent, Good, Fair, or Poor' and the respondent gives an answer of "Pretty Good" then you must say something like "I see, so would that be Excellent, Good, Fair, or Poor" It is important that you re-read the entire scale for the individual so that they can pick a valid answer choice. It is impossible for you to determine what the respondent meant by 'Pretty Good' so you must re-read the scale.

    7. Finish the interview:

      8. Conclude the interview with a genuine thank you for the respondent - we want to send them off with a good feeling about what they just did. Again, emphasize how important the survey is and how seriously it will be taken by Ohio State researchers.

      Go to: Quiz for Training Module 2: Interviewer Training
      Table of Contents
      Module 3: Survey Content