Research for Global Development

Six Tips for Face to Face Field Interviews


In eight years of conducting research across a number of countries in sub-Saharan Africa, I observed a lot of fieldwork, mostly face-to-face, paper-administered interviews. Year after year, the same mistakes are repeated, despite intensive training of field teams by experienced trainers. I observed and participated in some great classroom trainings where it appears that everyone understands how to proceed.  However when the team gets on the ground for the actual work, the same common errors frequently affect the quality of interviewing. My recent trip to Tanzania confirmed previous experiences.  While no interviewer will ever be perfect, these six tips should help avoid the most common errors I have witnessed:

  1. Be patient when asking multi-response questions.  Questioners must allow enough time for the respondent to exhaust all the potential answers, allowing for detailed probing that doesn’t rush the respondent. This is especially important when administering multi-response questions that require the interviewer to read multiple responses.  Be sure to get a response after reading each attribute before going to the next. Do not read all the attributes and tell the respondent to select the ones that apply.  The likelihood of remembering the first attributes is very low unless the list has less than five items.
  2. Correctly Use Show Cards. Only hand over the show card to the respondent once at the specific question requiring a show card. After the respondent has answered the question, take back the cards. Do not leave the show cards with the respondent when the interview is in progresses – many respondents will start to flip through the pages and not concentrate on the questions being asked. In case the respondent cannot read or is having difficulty reading items on the show card, please read out the items on the show card for him. Ensure that answers based on the show card are exactly as they appear on the cards e.g. if the cards have answers like, ‘trustworthy very much’ and the respondent says ‘like very much’ (which is not one of the answers on the card)-these are two totally different responses. In such a case get the respondent to look at the responses on the show card and pick an answer from one of the options given. All show cards should have card numbers for ease of execution for the interviewer.
  3. Always read the question as they appear on the printed page.  Questioners should not interpret or explain the meaning of the question to the respondent. Repeat the question if the respondent does not understand, or is unclear. Only explain terms that are not clear to the respondent.
  4. Never allow the respondent to read the content of the questionnaire.  Unless it is a self completion questionnaire, questioners should sit in a position that does not allow the respondent to glance at the content or even see how answers are coded. If you have to shift your sitting position, make a kind and tactful explanation to the respondent.
  5. Transition smoothly. When a questionnaire moves into a new section – or there is a change of thought or line of reasoning in the questionnaire – make a short introductory statement that helps the respondent adjust his or her thoughts to the new section.
  6. Listen carefully to the respondent in order to avoid asking extra questions that have already been answered. If responses do not fit in a pre-coded list, simply utilize the ‘other response’ code.

These common mistakes have a big impact on quality of data collection. Such errors can water down the results of even the most rigorous sampling techniques of household and respondent selection.

My next field trip is coming up soon and I will keep you posted on how it goes!

InterMedia

Six Tips for Face to Face Field Interviews


In eight years of conducting research across a number of countries in sub-Saharan Africa, I observed a lot of fieldwork, mostly face-to-face, paper-administered interviews. Year after year, the same mistakes are repeated, despite intensive training of field teams by experienced trainers. I observed and participated in some great classroom trainings where it appears that everyone understands how to proceed.  However when the team gets on the ground for the actual work, the same common errors frequently affect the quality of interviewing. My recent trip to Tanzania confirmed previous experiences.  While no interviewer will ever be perfect, these six tips should help avoid the most common errors I have witnessed:

  1. Be patient when asking multi-response questions.  Questioners must allow enough time for the respondent to exhaust all the potential answers, allowing for detailed probing that doesn’t rush the respondent. This is especially important when administering multi-response questions that require the interviewer to read multiple responses.  Be sure to get a response after reading each attribute before going to the next. Do not read all the attributes and tell the respondent to select the ones that apply.  The likelihood of remembering the first attributes is very low unless the list has less than five items.
  2. Correctly Use Show Cards. Only hand over the show card to the respondent once at the specific question requiring a show card. After the respondent has answered the question, take back the cards. Do not leave the show cards with the respondent when the interview is in progresses – many respondents will start to flip through the pages and not concentrate on the questions being asked. In case the respondent cannot read or is having difficulty reading items on the show card, please read out the items on the show card for him. Ensure that answers based on the show card are exactly as they appear on the cards e.g. if the cards have answers like, ‘trustworthy very much’ and the respondent says ‘like very much’ (which is not one of the answers on the card)-these are two totally different responses. In such a case get the respondent to look at the responses on the show card and pick an answer from one of the options given. All show cards should have card numbers for ease of execution for the interviewer.
  3. Always read the question as they appear on the printed page.  Questioners should not interpret or explain the meaning of the question to the respondent. Repeat the question if the respondent does not understand, or is unclear. Only explain terms that are not clear to the respondent.
  4. Never allow the respondent to read the content of the questionnaire.  Unless it is a self completion questionnaire, questioners should sit in a position that does not allow the respondent to glance at the content or even see how answers are coded. If you have to shift your sitting position, make a kind and tactful explanation to the respondent.
  5. Transition smoothly. When a questionnaire moves into a new section – or there is a change of thought or line of reasoning in the questionnaire – make a short introductory statement that helps the respondent adjust his or her thoughts to the new section.
  6. Listen carefully to the respondent in order to avoid asking extra questions that have already been answered. If responses do not fit in a pre-coded list, simply utilize the ‘other response’ code.

These common mistakes have a big impact on quality of data collection. Such errors can water down the results of even the most rigorous sampling techniques of household and respondent selection.

My next field trip is coming up soon and I will keep you posted on how it goes!

Marketing Materials

Contact Us:

InterMedia Headquarters

1825 K Street, NW
Suite 650
Washington, D.C. 20006
+1.202.434.9310
FAX: +1 202 434 9560
Contact | View Map

InterMedia Africa

UN Avenue, Gigiri Nairobi
Box 10224
City Square 00200
Nairobi, Kenya
+254.720.109183
Contact | View Map