COSP Home
Archives   Calendar    Coordinating Center    Multisite Activities    
Project Description
  COSP References  Study Sites    Search Our Site   
Table of Contents
    Upcoming Meeting Agendas

______________________________________________

 

V. Asking the Questions

As an interviewer, you must be aware of everything that is going on during the question and answer process.  The following information should be used as a guideline to help you to recognize potential problems during the interview.

A.      Interviewer Effects:

The interviewer can influence the respondent’s answers consciously or unconsciously through the use of verbal and non-verbal cues. You must avoid interjecting your own expectations and values that could lead the respondent to provide biased answers.  Respect the respondent’s personality, customs, and cultural background and do not impose your own beliefs, values, and interpretations on the respondent.  Always maintain a neutral approach and do not distort the wording of questions or instruction guidelines.  If a respondent tells you about a painful experience to which you relate, do not share your own experience with them.  Instead, you may say, “I am sorry that happened to you,” or “That must have been a difficult time for you,” or you may offer a contact name and telephone number from the resources list in the Appendix.

To avoid creating interviewer effects, do not:
bulletOffer your own opinion during the interview
bulletDisplay approval or disapproval through your tone of voice, facial expression, or side comments
bulletDiscuss your own experiences with the respondent
bulletRead the questions using your own words instead of those written on the questionnaire

Examples

The following examples show situations in which you could influence the respondent:

 

1.       A question reads, “What is your profession?”

The Interviewer asks, “What is your current job?”  A teacher by profession might be currently working in a grocery store because of a teacher’s strike and would answer the first question “teacher” and the second “grocery store clerk.”  The correct response is lost, and what’s worse, those who interpret the data will never know it unless a supervisor has observed the interview. 

2.      A question reads “How did you find out about our program?” with Interviewer instructions, “Do not read response options.”

The Interviewer does not read the whole list of options out loud but begins offering some of them when the respondent hesitates. The respondent was about to say she was told about the program by a friend, but when the Interviewer suggests a TV commercial, she says, “Oh, maybe I did see one.”  She does not go on to say that her friend’s recommendation is what most motivated her to look into the program. The correct response has again been lost. 

                    3.       A question reads, “What is your opinion of how well the President is doing his job?”
Although the Interviewer’s probing instructions are to remain neutral, to say “un-huh” and “please continue” to get a complete response, when the respondent says he is happy with the President’s performance, the Interviewer chuckles and asks, “Well, what about that illegal arms deal incident?” The chuckle tells the respondent that the Interviewer disagrees with him or her, and the question about the arms deal takes the respondent in a direction he would not have chosen if left to respond on his/her own.

(From Frey and Oishi, How to Conduct Interviews by Telephone and In Person, Sage, 1995, p. 34)

B.     Listening Skills:

In this context, listening involves two important aspects.  First, by listening attentively to the respondent you will be able to probe for more information when necessary and be sensitive to the respondent’s level of comfort or discomfort with the question.  Second, you must be able to hear and understand the respondent’s answer to record it correctly on the questionnaire.  The best conditions for active listening include a state of “relaxed concentration” during which the interviewer is listening to the respondent, watching the respondent’s face and body language, and trying to understand the respondent’s ideas.  

C.     Body Language/Non-verbal Cues:

As an interviewer, you must be aware of your own and the respondent’s physical expressions.  Non-verbal cues can be conveyed through facial expressions, posture, and hand and foot movements.  Slouching in the chair, resting your head on a hand, tapping your fingers, shaking your foot, playing with a pencil, or other object, yawning, fidgeting, or staring may all signal lack of interest in, anxiety about, or frustration with the interview process.  Other body signals may indicate different emotional reactions, including sadness, anger, or confusion.  Observe the respondent’s body language when he or she first enters the room and during the course of the interview.  If the body language changes and there are obvious prolonged signs of distress during the interview, you may need to offer the respondent a short break to get a drink, go to the restroom, or walk about.  Also be tuned-in to your own body language.  You may be unintentionally telling the respondent you are bored, uninterested, or impatient, and this could affect his or her attitude and answers.    

D.    Reading the Questions:

When beginning an interview, you should reassure the respondent that there are no wrong or right answers.  As an interviewer you must:

 

bulletRead questions exactly as they are worded in the questionnaire.
bulletRead questions in the order in which they are presented on the questionnaire.
bulletAsk every question on the questionnaire (unless there are skipped instructions).
bulletAfter reading the question, listen quietly and patiently for the response.
bulletDo not interrupt or make a comment before the respondent has completed an answer.
bulletRead questions with no additions, deletions, or substitutions.
bulletRead each question slowly at about two words per second.
bulletUse a tone of voice that conveys assurance, interest, and a professional manner that is neutral and non-judgmental.
bulletEmphasize underlined words to enhance meaning.

Do not attempt to re-word or explain a question.  If the respondent does not understand, repeat the question slowly.  You can also use the standardized definitions presented with the questionnaire (see Section IV).  If they continue to ask what it means, reply, “Whatever it means to you (MTY).”  

Sometimes a respondent will want to answer a question before you have finished reading it.  It is crucial that the respondent hear the entire question before answering.  If the respondent interrupts, you must continue reading the question.  This allows the respondent to hear the entire question and it also discourages future interruptions.  

Don’t skip a question because the answer was given earlier or because you “know” the answer.  In those situations in which the respondent has already provided information that probably answers the next question, you may preface the question with some combination of the following phrases: 

bullet“I know we’ve talked about this,”--or—“ I know you just mentioned this, but I need to ask each question as it appears in the questionnaire.”
bullet“You have already touched on this, but let me ask you…”
bullet “You’ve told me something about this, and this next question asks…”

Do not direct the respondent toward an answer or assume that an “answer” you got in passing is the correct answer to a specific question at a particular point in the interview.  Do not direct the respondent by mentioning an earlier answer.  If an answer is different from the one you expect, do not remind the respondent of an earlier remark or try to force consistency.  Finally, remember that although you may have read these questions many times, the respondent is hearing them for the first time and needs time to understand the questions in order to decide on the answers.  

E.     Using Interview Probes:

There may be situations in which the response to the question is unclear, incomplete, or not related to the question.  Some respondents may frequently reply that they “don’t know” the answer.  In these situations, you can use an interviewing technique known as the “probe.”  The probe is used to obtain more information, but please remember that probing must not bias the respondent’s answer.  Do not supply an answer for the respondent by probing them to elicit a specific answer.  Here are examples of probes that are permitted in an interview:  

 
bulletShow Interest. An expression of interest and understanding, such as "uh-huh", "I see", or "yes", conveys the message that the response has been heard and more is expected.
bulletPause. Silence can tell a respondent that you are waiting to hear more.
bulletRepeat the question. This can help a respondent who has not understood, who has misinterpreted, or who has strayed from the question.
bullet Repeat the Reply. This can stimulate the respondent to say more, or to recognize an inaccuracy.
bulletAsk a Neutral question. "Can you tell me more about that?"
bulletFor Clarification: "What do you mean exactly?", or "Could you please explain?"
bulletFor Specificity: "Can you be more specific about that?" 
bulletFor Relevance: "I see. Well, let me ask you again." (REPEAT QUESTION AS WRITTEN)


(From Frey and Oishi, How to Conduct Interviews by Telephone and in Person, Sage, 1995, pp. 123-124) Go to Exercise #3

Probe Abbreviations:

 

Repeat question                         (RQ)

Repeat Frame of Reference  
(RFR)

Repeat Choices
(RC)

 

Whatever ______ means to you

Whatever you think of as ________________

What do you mean?     (WM)

How do you mean

Would you tell me more about your thinking on that?                 (TM)

Would you tell me what you have in mind?  

What do you think? (WT)

What do you expect?

Which would be closer to the way you feel? (WC)

Are there any other reasons why you feel that way?           (AO)

NEUTRAL PREFACES TO PROBES THAT SHOULD NOT BE RECORDED:

Overall…                                                    

Generally speaking…

Well, in general…

In the country as a whole…

Yes, but…

Of course no one knows for sure…

Of course there are no right or wrong answers…

We all hope, but…

We’re just interested in what you think…

Let me repeat the question…  

1.      Use of Probes

 

It is important to read the questions slowly, pausing when necessary to allow the respondent time to provide complete and accurate responses. At first it may be difficult to know when to repeat questions or how to long to pause, but this will become easier with practice.  Repeat the entire question if the respondent indicates that s/he did not understand it.  In contrast to those who are not providing enough information, there may be respondent’s who talk excessively or ramble in their replies.  You will have to remind them that it is important to complete the interview in the given time.  You may also have to refocus some respondents by saying “Let me make sure I have this down right” and repeating the answer to bring them back to the survey process.

 

What about the respondent who gives an “I don’t know” answer?  Consider what s/he might really mean.

bulletRespondent doesn't understand the question but doesn't want to admit it.
bulletRespondent is thinking about his/her answer and is filling the silence.
bulletRespondent doesn't want to answer the question.
bulletRespondent really doesn't know or doesn't have an opinion on the subject.

Remember: You should probe a “don’t know” response at least once.

The most effective probe for a “don’t know” is to repeat the question (RQ) or pause.  

bulletExercises #4-8

F.      Displaying the Scale Cards:

There are a number of questions on the survey form that have more than 4 or 5 responses.  It would be too difficult for the respondent to remember all of the choices.  For this type of question, a scale card is supplied with the reply choices listed on it.  Each of these questions will tell you which scale card to use.  Before you read the question, hand the card to the respondent.  Read the question, ask the respondent to look at the code card as you read the responses, and ask the respondent to choose one answer from the list.  You will use a code packet, so that you can flip to the appropriate card to display to the respondent.   

G.    Answering Questions From the Respondent:

During the interview, the respondent may have questions about the purpose of the survey, the meaning of the questions, or about you as the interviewer.  Remember that you must try to keep the answers to the questions as standardized and unbiased as possible.  For that reason, you must not engage in a general conversation or lengthy explanations.  For questions regarding the survey process, use the standardized replies provided in section IIIE.  Do not give personal information to the respondent, because you will be conducting other interviews in this setting and may also be doing a re-test interview.  

Other key phrases that may be used when the respondent asks for more information include:  

bullet“This is all the information available to us.”  
bullet“We would like you to answer the question in terms of the way it is stated.  
 
bullet“Could I read it again for you?”  
bullet“I’m sorry, I don’t have that information.”     
bullet“I will write on the questionnaire the qualifications to your answer that you have just mentioned.”

If the respondent still requires more information, instruct them to contact the project director.

(From Frey and Oishi, How to Conduct Interviews by Telephone and in Person, Sage, 1995, p. 134)  

H.  Feedback:

Feedback consists of statements or actions that indicate to the respondent that s/he is successfully answering your questions, and encourages appropriate behavior and responses for future questions.  Feedback should not be confused with interviewer effects.  Giving the respondent feedback does not lead anyone to answer in a specific way, but rather it indicates to the respondent that the answers they are providing are appropriate for the questionnaire.  You must decide when and how to give feedback to the respondent. Respondents who exhibit appropriate behavior will listen to the entire question before beginning to respond, give serious and clear answers, and stay on task when answering a question and avoid digression to unrelated stories or topics.  If a respondent interrupts your questions, offers partial or unclear answers, or answers each question with a story, this is inappropriate and you should use probes or neutral non-verbal techniques to encourage more appropriate responses.   

Since you are interviewing in person, you will have the opportunity to use non-verbal feedback.  “Small talk” before beginning an interview may help to establish trust between you and the respondent.  Once the interview has begun, a smile, nod of the head, or eye contact will acknowledge the respondent’s answer and encourage similar responses.  However, non-verbal feedback must be used carefully in order to encourage appropriate behavior or discourage inappropriate behavior.  For example, if a respondent refuses to answer a question or strays from the topic, you must not smile or nod your head.  If this should occur, use probing techniques or ask the respondent a neutral question.  However, please remember that while it is important for interviewers to be objective, it is essential that they don’t come across as cold.  Interviewers should be both professional and friendly, but this balance can only be achieved through practice.     

FEEDBACK PHRASES

Short

bullet

I see...

bullet

Uh-huh/Um-hmm.

bullet

Uh-huh/Um-hmm, I see.
 

bullet

Thank you.  

bullet

Thanks.  

Long

bullet

That’s useful/helpful information.

bullet

It’s useful to get your ideas on this.  

bullet

Thanks, it’s important to get your opinion on that.  

bullet

I see, that’s helpful to know.  

bullet

It’s important to find out what people think about this.  

bullet

That’s useful for our research.  

Interviewer Task-Related Comments

bullet

Let me get that down.  

bullet

I need to write it all down.  

bullet

I want to make sure I have that right (repeat answer)  

bullet

We have touched on this before, but I need to ask every question in the order that it appears in the questionnaire.

 

For further clarification on feedback, read the following examples and then examine the page of Feedback responses:  

Example #1:

 

Interviewer: Do you do any volunteer work or any other kind of work for which you are not paid?

Respondent: Yes, I really like to volunteer at my church, by offering to organize events.

Interviewer:  Thank you.  That was a very thorough answer.

 

Example #2:

Interviewer:  What is your current marital status? Married, widowed, divorced, separated, or never married?

Respondent:  I am married but living separately from my husband and children this year.

Interviewer:  I see.  Now, what is the highest level of school that you have completed?

   
In the examples above, “I see” and “Thank you” are feedback phrases used to indicate that the respondent is providing appropriate answers.  In contrast, the next example demonstrates feedback to an inappropriate answer:

Example #3:

Interviewer:  Have you had any problems associated with alcohol use in your lifetime?

Respondent:  Well, growing up, my Aunt Judy used to tell me that I was always going to be a good drinker, then when I went to college, I would hang out with my friends, and drink every now and then, but to this day I still don’t like the taste of alcohol, so I guess Aunt Judy was wrong.

Interviewer: I see, (pause) but, ‘Have you had any problems associated with alcohol use in your lifetime?’

R:  I guess not, no.  

Remember that effective Interviewers give feedback for good performance, not “good” content.  Interviewers should not use the phrases, “O.K.” or “all right” when providing feedback to a respondent because they indicate agreement with the response.  Instead, you should always use neutral phrases that do not indicate agreement.  As a general rule, you should give short feedback phrases for short answers and long feedback phrases for longer answers.  You can also use a brief pause followed by a feedback phrase to make it more powerful.  The pause signals to the respondent that you have considered his/her answer carefully.  An Interviewer should determine how often to give feedback by considering the performance of the respondent.  Some respondents may need more feedback to encourage appropriate responses while other respondents may need less feedback.  

bullet

Go to Exercise # 8  

 

I. Recording the Responses:

The respondent’s answers must be completely and properly coded on the questionnaire or in the laptop, or the interview results cannot be used for analysis.  Record every answer in the appropriate category.  If a response has been “Don’t Know,” enter NANS for that question.  If the respondent refuses to answer the question, also enter NANS.  If you must probe for a response, write P next to the question.  If the respondent has difficulty in understanding the question, make a note next to the question.  You may also use the interviewer’s comment section for further explanations.  Every question must have some recorded answer, or an explanatory mark, in the available space.  If an item was part of a skip pattern, or if you forgot to ask it, enter NASK for that question.  Common errors made by the interviewer include:

(1)    Omitting an answer.

(2)    Recording the wrong answer code.

(3)    Circling more than one answer or entering more than one number.

(4)    Writing illegibly, or in abbreviations that are not readily understandable to the people who are coding.

(From Ralph, “Research Interviewer Handbook,” p. 7)  

If you accidentally mark the wrong answer, cross out your mark, write ‘error’ next to it and circle the correct answer.  If you are using a laptop, delete the previous answer and enter the correct answer. Some respondents may change their mind while giving an answer, so it is useful to wait a couple seconds between questions to be sure the respondent has finished speaking before marking their final answer.  If you have finished the interview, but have some concern about whether the respondent answered accurately or truthfully, make a note of this or of any other concern in the section for Interviewer’s comments.  

J.    Taking breaks

Half way through the interview, you should take a 10-minute break. This gives you and the respondent a chance to stretch your legs and rest your minds.  You will get more accurate and complete responses during the last half of the interview if the respondent has had a chance to rest half-way through. Some respondents may need more than one break.  Be aware of fidgeting or other non-verbal behaviors that indicate that the respondent is tired or restless.  Be careful during the break not to talk about anything that would bias answers to any of the rest of the questions in the interview.  Small talk such as sports or the weather are good topics of conversation during the break.    

K.  Troubleshooting

It is rare but possible that a person who has agreed to be interviewed may become uncooperative during the interview causing you to feel uncomfortable about the respondent’s behavior.  If you feel that the person cannot actively participate in the interview because he/she is uncooperative, rude, threatening or hostile towards you, or responding in a sexually charged manner, you should end the interview immediately.  Conducting interview’s at a neutral site, such as the traditional service agency or drop-in center, can reduce these types of behaviors. Therefore, during this project, all interviews will be conducted at the [traditional service agency or drop-in center].  If an alternate interview setting is needed for a respondent, notify the Site Coordinator to receive approval. This policy is intended to ensure your safety and should be followed at all times.

Although we will be using a neutral site to conduct interviews, it is still possible for people to act inappropriately.  Since every person has a different comfort level, it is impossible to foresee all situations that may cause Interviewers discomfort.  As a general rule, Interviewers and respondents should refrain from the following:

bullet

Touching other than to shake hands

bullet

Using profanity

bullet

Acting out (yelling, slamming furniture). 

You should also be aware of other behaviors signifying that the respondent is not able to complete the interview, including signs indicating a psychotic episode or illiteracy.  If at any time you experience discomfort due to the behavior of the respondent, you should end the interview immediately.  Tell the respondent that you have finished your questions, that you appreciate his/her time, and that if the respondent has any questions they should be referred to the Project Director(s) [List site director].  Then promptly leave the interview site.  Make written notes in the post-interview Observation Section, and then report your experience to the Field Director immediately, before you conduct another interview.  If this kind of problem should occur, please stay calm, and know that their behavior is not related to you or this project.  Always remember that your safety and comfort comes first and that if you have to end an interview, this will not be held against you. 

L. Ending the Interview:

Once the interview is completed, thank the respondent for his/her time and effort.  Let the respondents know that their participation was very important for the success of the project.  At this time, they may express some concern about the content of the survey questions and you may need to spend a short time reviewing the purpose of the survey and use of the information.  If necessary, give the respondent the site coordinator or project director’s number and reassure the respondent that he/she may contact the them at any time regarding the survey content or process.  Remind respondents of the payment schedule and have them sign the [Payment Reimbursement] Form (forms are in the Appendix).  Instruct respondents that if there are any problems in receiving the payment, they should contact the field coordinator.  

M. Post Interview Observations:

Once you have finished the interview and left the interviewing site, take a couple minutes to make some written notes about the interview and the respondent.  Record any problems you may have had during questioning, whether the respondent seemed attentive and responsive, and whether the respondent understood the questions.  It may be useful to record specific behaviors, statements and impressions that occurred during the interview.  In addition, you should record any errors or mistakes that were made during the interview.  These notes will help you with future interviews and will also help the Project Director ensure the validity of the project by noting respondents who may have been giving false or biased answers.  This process can also help other interviewers because you can later draw upon these notes to share your experiences with each other.      

Back ] Up ] Next ]

Missouri Institute of Mental HealthBullet5400 Arsenal StreetBulletSt. Louis, Missouri 63139
BulletPhone: 314-644-8787 BullletFax: 314-644-8834