11  Define and diagnose the problem

Before getting to the trial, you will typically have already done much work. Primary among these is defining and diagnosing the problem.

Defining and diagnosing the problem is a critical step in good intervention design. This is not the focus of this subject, but you need to clearly define the problem, where you should focus your attention, what particular outcomes you care about the most, and who your target population is. You must also be mindful of your resources and consider what is realistically achievable. We must have a thorough understanding of the context. What is the reality on the ground? What problems and opportunities are the targeted people facing?

Descriptive methodologies can be very powerful. They are the basis of good intervention design. The understanding of the context that they provide helps ensure that an intervention is designed to solve a problem that exists in a specific context, matters to the targeted people, and fits their constraints and opportunities.

One methodology for defining and diagnosing the problem is a needs assessment.

Needs assessments aim to generate a description of the problem and existing solutions using interviews and focus groups, existing data, or new surveys. Needs assessments can also help pinpoint weaknesses in existing programs.

The types of questions needs assessments can answer include determining what groups should be targeted by the program (for example, what groups have the greatest need or would benefit the most) and what are their problems and opportunities. It is also important to understand the possible reasons for the problems they face and what they are already doing to resolve these problems. Finally, it is essential to understand whether existing programs are already focusing on these underlying needs and what challenges remain unaddressed.

To see how this works with a concrete example, let’s look at the questions that Glennerster and Takavarasha (2013) seek to answer with a needs assessment before developing a new educational program for primary school.

If we are planning an education program (or an evaluation) for primary schools, we need to have a good understanding of the level of learning among primary-age children in the area.

Who is falling behind, and who is not? What are the possible reasons that learning levels are low? What are the child absenteeism rates? What are the teacher absenteeism rates? (If children who attend regularly are learning well, child absenteeism could be the problem.) Or are children who attend regularly still falling behind? What do parents and children say about why they do not attend and why they find it hard to learn? What do teachers say about why learning levels are low? How are classes conducted? What material is covered? Is it appropriate to the level of learning of most children? Are children sick or hungry? Can they see and hear what is going on? We also need to know if other providers are working in the area. What are they doing?

Glennerster and Takavarasha (2013)

A range of methodologies can be used to conduct a needs assessment. Focus groups can tell us that people might find a program is located too far away or operates at inconvenient times; absenteeism surveys can show us whether program staff are regularly absent; take-up surveys can assess what fraction of potentially eligible participants are actually using a service, and random supply checks can make sure that supplies are getting through. A combination of both qualitative and quantitative techniques is useful in gaining a good understanding of the context.

Further references on defining and diagnosing the problem include: