MONITORING AND EVALUATION QUESTIONS: THE RIGHT MIX

Today I would like to talk about having the right questions in MEAL whether at the level of conceptualizing the exercise or even during the seeking of information on the progress and performance of a project or programme. I have gone to this area because I get disturbed seeing long list of questions in both academic and programmatic work. People sort of "feel happy" seeing long lists of questions and I wonder if these same questions would auger well when posed to them as respondents. The answer is as good to you as it is to me. Why don't you to others what you want them to do unto you?

My article today therefore focuses on having the "right mix" of questions in MEAL. Remember that MEAL is about your gathering information to tell you where you are in your work so that you improve but so that for you to account your constituencies and donors. This means that you are not gathering data for data's sake. You have specific intentions for the information that you seek. Make these very clear and you are home and dry.

Once you are clear that you know why you want certain information or data, check if this data or information exists inside your data base or with someone/somewhere else/ I am saying this because often we rush to collect information which is already gathered by others. It a waste of the time for the communities giving us this information. In addition, it adds to the confusion and duplicity of development organizations within organizations. This could be another subject matter altogether. To make matter worse, this only leads to information overload because hardly is all the data adequately used.

Assuming there is no available information or not all is required, you would then proceed to draft questions on the areas for which you require data. The cardinal rule taught in many diploma and undergraduate Research Methods classes are as follows:

• before writing the questions, ensure to have a clear and succinct introduction for building trust and understanding
• use a thematic approach to develop the questions
• start with the simple moving to the more complex questions
• for ease of analysis, pre-code (give alternatives to) the questions
• for each pre-coded question, e.g. "do you smoke" (a) Yes (b) No, add an open question, e.g. Please give reasons for your answer
• avoid leading questions, i.e. those that suggest answers, e.g. You smoke twice a day, do you?
• avoid complex or double faced questions, e.g. you smoke as you take alcohol isn't it?
• use simple and clear language
• provide the biodata space at the end. This used to come at the beginning but nowadays these details are being requested towards the end because they are very personal and are best requested at the end once rapport has been established. Obviously those personal details that are important for selection of the interview, e.g. head of household, must be sought as required.

I have listed few points on what to observe or not observe in writing your questions. These are not magic ideas but they need to be seen as reminders. There are other considerations in MEAL like the often used DAC criteria (https://www.oecd.org/dac/evaluation/daccriteriaforevaluatingdevelopmentassistance.htm) and these aspects needs to be factored. However, it is useful to be creative to know who to ask certain questions on these criteria as opposed to seeking same information to every respondent.

In summary, the consideration in MEAL is on seeking appropriate information without tiring respondents. They need to respond to your questions and go work. Organizations have been accused of being "Lords of Poverty". Do not fall into this category by taking so much time with people responding to your questions that they have no time to go work and remove themselves from poverty. Take not more than an hour of their time and release them to do their normal business.

Enjoy your week and work smart.