Questioning
In research projects
Descriptive question, used primarily to describe the existence of some thing or process.
Relational question, designed to look at the relationships between two or more variables.
Causal question, designed to determine whether one or more variables causes or affects one or more outcome variables.
In surveys
Dichotomous questions, usually these questions require yes/no answers or require a person to answer by choosing an option(s) from a multiple choice of possible answers.
Nominal questions, these
types of questions are designed to inquire about a level of quantitative
measure. Usually these questions form correlations between a number and a
concept. For example:
Occupational Class: 1= Moderate 2= Severe 3= etc. [2]
Qualifying questions (a.k.a. filter questions, or contingency questions. These types of questions are designed to determine if the individual answering the question needs to continue on to answer sub sequential questions.
----
Why is it important to ask ‘good’ questions?
The questions we ask, how we phrase them and the even the tone that we use can affect the quality of the answers that we get. Good questions prompt the person being questioned to think about all the impacts that an intervention has had on their lives, and to give full and open answers.
Good questions are usually:
1. Simple
Good questions should be imaginative and thought provoking. But the more abstract a question is, the more potential there is for misunderstanding. Try to find the right balance. Avoid technical jargon, use terminology and language appropriate to interviewees.
Asking people to imagine a different situation can be an effective way of exploring their views. (eg “If your daughter could go to school, what difference do you think this would make to her future?” or “What do you hope that your life will be like in the future?”). But remember that for some people it may be difficult to imagine alternatives when they have not had the luxury of choice before. In these cases, starting from the present and the practical may be more effective – eg “What needs to change to solve the problem you are currently facing?”.
Asking people to compare and contrast situations (before and after, different approaches etc) is a good way to explore the impact of an intervention. But phrase them carefully; multiple questions may confuse the person you are questioning and lead to valuable answers being lost. Break the question down into simple components and ask each one as a separate question.
2. Appropriate
Good questions build from an understanding of the frame of reference of the person we are questioning, and their cultural norms. For example, is it ok to ask direct questions? Are there gender issues to consider? An apparently straightforward question such as “How old are you?” or “What time do you wake up?” can be meaningless in a context in which people do not record birth dates or have clocks.
Without a good understanding of the frame of reference of the person we are questions, we may miss key information. For example, a health worker in Latin America was unable to understand why villagers did not boil their water after she had spent many months explaining that this would kill germs which cause diseases. A later evaluation, asking more appropriate questions, found that the villagers had simply not believed her ‘story’ about the germs and that, traditionally, they only gave hot drinks to people who were already sick. Therefore her questions did not make sense to them. Questions that started from their understanding and their beliefs were more effective in ascertaining the facts.
Inappropriate questions may not only mean we miss key information, they may also mean we inadvertently cause insult to the person we are questioning. In some African countries it is considered bad luck to state how many children a person has, so it would be insensitive to ask. Get local advice on what taboos there might be and how to phrase difficult or personal questions.
3. Open
A closed question is one that allows the person you are questioning to only answer yes or no. For example, “Your life improved a lot since you received these goats, hasn’t it?”. With such a question you have in effect provided the answer yourself, and are merely seeking confirmation or denial. But the truth may be more complex, and by only allowing a yes or no answer, you will miss this complexity.
Rephrasing it as an open ended question such as “Since you received these goats, what differences have you noticed in your life?” would allow the person to answer something like; “I have more income as I can sell the milk, but I now have a lot more work as I have to clean their pen and fetch more water. It was easier when I had chickens, they are less work but can earn me the same amount of money.”
Closed questions may also be ‘leading’ questions; ie they lead the person you are questioning to give a particular answer (eg “My life has improved”). You will only find out the real impacts – good and bad - of the intervention if you allow the person you are questioning as much space as possible to develop a full and honest answer, which may include things that you didn’t expect.
NB There may be circumstances in which it is appropriate to ask closed questions, for example when canvassing opinions from a wide range of people or when trying to ascertain a very simple fact.
4. Use “What?” and “How?” rather than “Why?”
Often we need to find out the reasons for things or for people’s opinions, and so we ask “Why?”. But the question “why” can make people feel defensive and can create misunderstanding. There are other, more effective ways of finding out the reason for things… including simply asking “What made you decide to do it this way?” (not “Why have you done it this way?”). Or “What makes you feel that?” (Not “Why do you say that?”).
“Why” also assumes cultural norms. What one person considers a perfectly adequate reason for doing or saying something can be very different from what another person is trying to find out. By keeping the question focused on what you really want to find out, you are more likely to get the answer you are looking for. For example, NGO officers travelling to a project in a desert region in India noticed that the older women wore heavy, black woollen clothing. They discussed why this might be, eg did the wool and the dark colour deflect the heat? On arrival, they asked the women, who replied with an epic story including the mythological origins of their practice of dressing as they did. The question “why” had different meanings for the villagers and the NGO staff, the former gave cultural reasons and the latter sought practical reasons. If the NGO had wanted a practical answer, they could have asked a more practical question such as, “Are those clothes cooler?”
Questioning and listening: see also Listening and Feedback
Interviews are usually best when conducted like conversations, rather than interrogations; a conversational tone can make people feel more relaxed and will motivate them more to give full and informative answers.
However, if the person is a senior government official or community leader, it may be important to show some deference for their status by not seeming too ‘casual’. Of course, all individuals, regardless of their social status, should be treated with respect and should know that we understand that their time and opinions are valuable.
The way that you hear, listen to and respond to the answer will also have an impact on the accuracy and fullness of the information you receive.
Being a good listener means being completely open to what you are hearing and showing that you are paying attention and trying to understand. Do this by maintaining eye contact (if culturally appropriate), even if using an interpreter, nodding and saying yes as appropriate, but avoid participating in the conversation excessively. Accept whatever is said with empathy and without judgement. Be especially careful not to betray your feelings if you are shocked by or disagree with what you hear, as this may inhibit the rest of the interviewee’s answers.
Be wary of only hearing what you want to hear, or of discounting information that you don’t understand. Ask for an explanation if you need to, and then repeat what you have heard back to the person so that you can be sure you have understood correctly. If you still can’t understand, say so in your notes.
Other information:
· Interviewing Key Informants