By the time most of us are adults, we know how to do a lot of things we take for granted. We’ve learned how to walk, ride a bike, drive a car, use a credit card, etc. Yet there are some things that seem so simple we think we’re expert at – even when we’re not. We know how to ask questions, don’t we? You simply realize something you want to know and then attach a question mark at the end. Sounds simple, right?

When you want accurate information, however, it’s not so simple. Here’s an example from my own experience to illustrate one of the possible ways we can influence the answers without even being aware.

When I was an intern at an ad agency, I went along to observe some in-home interviews with farmers about farm equipment. The interviewer I was observing (not usually a moderator) was told at the last minute to ask a couple of questions about farm-tires for the Goodyear account. The interviewer told me on the way to the interviews that she did not think she’d get much from the farmers about tires, but she had to ask. As we sat through the first two interviews, I watched her ask the farmers – “How important is it to you what brand of tires you buy for your tractors?” and both farmers sort of shook their heads and said they guessed not real important. But I was watching her, and as she asked the question, she shook her head as if she was saying “no.” I pointed it out to her afterwards and she was shocked, but in the next interview very consciously did not shake her head again, and the farmer talked for a half hour about brands of tires and how critical it is to get good ones – they are large and very expensive and last better!

This interviewer was completely unaware that she was communicating her expectations to the farmers she was interviewing – but if there had been no intern to watch her, the Team would have never known and would have come back with very different conclusions.

There is a large body of social-scientific research on interviewing and the influence the interviewer and the questions can have on what we learn. There are several ways that an interviewer can unconsciously bias their findings, which is why it is so important to hire an independent and experienced interviewer when conducting research. This should minimize the risk of leading the interviews and give the team members a chance to listen

· Leading questions:

We all know that open-ended questions are critical to getting good qualitative learning. But even among open-ended questions, there are ways to lead that should be avoided.

o Order of questions – one must keep in mind the effect of previous conversations on future ones: – make sure the conversation flows in an order that keeps people free to share.

§ Social desirability and pressures should be considered.

· For example, don’t have a conversation about the perceived dangers of germs in the kitchen before learning about someone’s kitchen cleaning habits – they might feel pressure to avoid telling about their less than pristine behaviors.

· Reverse that order, have them describe their behavior and then ask about germ-concerns.

§ If specific probes or diagnostics are asked before the general discussion (either in the interview or in a pre-interview questionnaire or screening call) people, including kids, often take their cue that these elements are the ones the interviewer wants to hear about. They might start talking about “texture” or “comfort of grip” when what is really important to them has been missed.

· How to avoid this: start broad, avoid diagnostics in pre-interview materials.

o Phrasing and wording of questions – Avoid language that has social-acceptability attached to it, for example, such as asking moms if they “care” about an issue with their kids would lead them to say yes, but asking them simply about the issue: tell me about ______________ would allow them to answer more honestly.

o Avoid putting words into the respondents’ mouths. -- When listening to friends, we often re-phrase what they have said or guess as to how it impacts them. In research interviewing, it is vital not to introduce language that is of interest… it should emerge from the interviewees.

o There are several types of leading questions*: Assumptive questions assume something in the question;

§ Linked statements is similar to order of questions in that what is said before the question influences the answer;

§ Implication questions imply cause and effect between two things;

§ Asking for agreement – suggest a desired answer

§ Tag questions are often yes or no questions tagged on to a statement

§ Coercive questions Imply trouble if the reply is not in a particular direction

· Body language that can influence

o Research shows that when interviewers take notes, respondents begin to give more answers like the ones the interviewer seems to be noting. Just note-taking alone can influence findings!

o Other body language, such as posture, nodding, smiling, frowning can have an impact on what people will share.

o How to avoid this? The interviewer needs to act. Interviewing is playing a role of a naïve and curious person who has no idea what they are about to hear. Sometimes the act is to be skeptical, to get the respondent to “convince” the interviewer of something. Sometimes the act is simply to not laugh when it might embarrass a respondent, or to pretend to completely like the respondent when they have just said something offensive. But always acting in service of keeping the unbiased, deep sharing flowing.

· Interviewer expectations

o A lot of research has been done on the impact of the expectations of the questioner. (Even with lab-rats, the researcher’s expectations can influence the findings!) Maybe you remember the research with teachers who were told which students were going to perform well and which were not, and then the fulfillment of these expectations despite the fact that the students were assigned randomly to the “expectation condition.”

o How does this affect interviews with kids (and adults)? If the interviewer has set expectations of what they will learn, they can easily give subconscious or unconscious cues to respondents, and nobody might even notice!

o What to do? It’s best to have an objective interviewer who does not know team expectations and especially team hopes about what they will hear. Interviewers/moderators need to work to stay in a place of curiosity rather than believing they already know what they will learn.

· Repeating questions

o Research shows that children will often change their answer if a question is repeated. Listeners sometimes interpret this to mean the respondent is wavering in their answer, and often dismiss the answers all together.

o However, a likely explanation is that kids (and adults too) can think there is a “right” answer – they spend a lot of time in school where this is true. So when a question is repeated, they try a different answer to get it “right”

o How to avoid this when you need to repeat a question? Try giving them permission to give the true answer. Something like: I am going to ask you something I already asked you – you did not get the answer wrong before, I just can’t remember what you thought about this. (I hesitate to use the word “wrong” – I’d probably remind them that there are no right or wrong answers at that point.)

*For more information on leading questions, see this webpage: http://changingminds.org/techniques/questioning/leading_questions.htm