Asking the Right Questions

An important prerequisite for crafting solutions is to ask the right questions.  Equally important is to whom the questions are directed and how the answers are processed.  Anecdotal answers, whether from experts in the field or those most affected by the problem are too variable and unreliable to be the basis for sound policy.  This concept is fully understood by the scientific community and by commercial marketing interests but is too often overlooked within the social sphere. 

For example, much current scientific research now entails mining existing data sources to find answers to new questions.  In evaluating the efficacy of a new drug the manufacturers must conduct a large scale drug trial to determine whether to continue production.  It would certainly not be acceptable to judge the value of the drug based on the verbal responses of the recipients alone, even if the numbers involved were large.  Although it is important to consider how people feel about the drug the key to success is to actually ascertain whether the drug does what it is designed to do and what, if any, serious side-effects occur.  Equally important is the inclusion of controls to determine the magnitude of the placebo effect (people often “feel” better just by taking a non reactive pill). 

Although companies do extensive marketing research by asking customers about their experiences they also pay handsomely for data analyses that indicate what their potential customers are actually buying because this is usually a better gauge of what they will buy in the future than a survey. 

In the same way, answering the vexing questions about the vagaries of human behaviour might be best served by formulating better questions and examining the existing data more thoroughly for the answers.  We know, for example that men do not access health services as regularly as women and thus their illnesses are often diagnosed late in the cycle and are more difficult and expensive to treat.  We can ask men about this and/or theorize as to why this might be the case but a better approach might be to thoroughly review all the data available on men’s access to health services to determine the answers to questions such as what eventually prompted them to seek medical attention, what times and days they actually first presenting in the system, whether they were encouraged to come by another person etc.  The answers to these questions would help to design appropriate public health messages and the opening hours of clinics.

We are confronted with evidence that although women are accessing education at all levels at a greater rate than men the gender imbalance in unemployment rates remains stubbornly high. Asking the right questions of the right people is probably a better route than asking individual women and girls to come up with reasons for this imbalance. Several questions immediately come to mind.  Firstly, an analysis of demographic data should tell us if the imbalance is different for women in different age ranges and educational attainment levels because that would show whether the problem is being addressed over time or not.  Another area of research should be to analyse whether there is a significant gender difference in the subject areas being selected by students.  All over the world there is a shortage of people with technical training and skills.  If analysis indicates that girls are under-choosing technical subjects then a response should include mechanisms to encourage more girls to select technical studies.  This might include better career counselling (for boys and girls) and making technical subject teaching more girl friendly.  

Belizeans are being encouraged to become more entrepreneurial and there is anecdotal evidence that women find it more difficult to access credit than men.  The first question to be asked is if this is in fact the case and the best people to ask are not individual women but financial institutions.  Banks and credit unions must have gender disaggregated records on applications for business loans and thus it should not be difficult to find out if there is indeed a gender imbalance in applications and/or in granting loans.  If the problem is that women are not applying for loans the best response would obviously be quite different than if they are applying and being refused.  If there is a gender difference in the distribution of loans the correct questions to the financial institutions should show if this is correlated with gender differences in collateral, business plans, other factors or a combination of these.  Only a comprehensive analysis will identify the heart of the problem and thus suggest the best methods for addressing it.
Data collection and data analysis may not be the stuff of headlines but if we do not ask the right questions we are unlikely to find the right answers.