M&E Fundamentals: Monitoring & Evaluation for development practitioners.

Reading · Core terms

Monitoring

The routine, continuous tracking of a project while it runs, to check that activities are happening as planned, on time, and reaching the intended people. Real-time. Helps you catch problems early.

Evaluation

The periodic, deeper assessment of whether a project worked and was worth it. Done at set points such as a baseline, midline and endline. Judges results, not just delivery.

M&E

The two working together. Monitoring watches the project as it happens; evaluation judges what it achieved. Both feed reporting and improvement.

MEAL

Monitoring, Evaluation, Accountability and Learning. The fuller frame now common in the sector. Accountability is showing, with evidence, that resources were used as promised, both upward to the donor and downward to the community. Learning is using the findings to make the next project better.

Project cycle

The path a project travels, from design through implementation to closure and the design of the next project. M&E belongs to every stage of this cycle, not only the end.

Baseline

A measurement taken at the start of a project, before any activity, recording the situation so later change can be judged against it. Sometimes you collect this yourself through a survey; sometimes you draw it from existing sources such as the Ghana Statistical Service or the district assembly.

Midline

A measurement taken partway through a project to check progress and allow course correction.

Endline

A measurement taken at the close of a project to judge what changed by the end.

Indicator

The specific, measurable sign you track to know whether something is happening or has changed, such as the number of farmers trained or the percentage reporting higher yields. Indicators are covered in depth in Module 3.

Reading · The results chain

A results chain is the causal logic of a project, in five links:

Inputs

The resources you bring: funds, staff, seed, transport, materials.

Activities

What you do with them: train farmers, set up demonstration plots, distribute seed.

Outputs

The immediate, countable products of those activities: 200 farmers trained, 10 demonstration plots established.

Outcomes

The changes in what people know and do as a result: trained farmers actually adopt climate-smart practices.

Impact

The long-term, higher change the outcomes build towards: yields, incomes and food security improve in the district.

Reading · A few more terms

Theory of Change

The explanation of how and why the change is expected to happen, including the assumptions the project depends on. It is the reasoning; the logframe is the summary table.

Logframe (logical framework)

A single table that sets out the project logic level by level, with an indicator, a means of verification, and the assumptions for each level. Indicators themselves are covered fully in Module 3.

Means of verification

Simply where the proof will come from: attendance sheets, KoboToolbox monitoring records, a baseline and endline survey, demonstration plot records.

Assumption

An external condition you are relying on but do not control, such as adequate rains, timely seed delivery, stable prices, or a community that stays settled.

Reading · What makes an indicator strong

A strong indicator is SMART: Specific (what, who, where), Measurable (a clear unit), Achievable (a realistic target), Relevant (it reflects the change you care about), and Time-bound (by when). If two honest people could disagree about whether it was met, it is not yet an indicator.

Reading · Indicators at every level of the chain

Output indicators

Count what you delivered: number of farmers trained, number of boreholes drilled.

Outcome indicators

Measure changed behaviour: share of farmers adopting a practice, share of households using safe water.

Impact indicators

Measure long-term change: household income, disease incidence, herds surviving a drought.

Reading · Two kinds, both needed

Quantitative

Numbers and percentages. They tell you how much and how many.

Qualitative

Descriptions of quality and perception. They often tell you why.

Reading · Three things that make an indicator usable

Baseline

The value before the project starts. Without it, you cannot prove change.

Target

The value you aim for by a set time, set honestly against the budget.

Means of verification

Where the data comes from: register, household survey, KoboToolbox records.

Reading · Primary and secondary data

Primary data you collect yourself (surveys, interviews, observation). Secondary data already exists (Ghana Statistical Service, district assembly records, health and school registers). Always look for secondary first; it is cheaper and faster than sending a team to the field for something already written down.

Reading · Match the method to the question

Survey / questionnaire

For reaching many people and counting. Answers how many, how much.

Key informant interview

For depth from a few knowledgeable people (the chief, the assembly member, the extension officer).

Focus group discussion

A small group together, exploring perceptions and reasons. Answers why, how do they feel.

Observation

Seeing with your own eyes whether something is really there and working.

Document review

Reading records that already exist (attendance sheets, health registers, enrolment books).

Reading · Routine monitoring, in one line

Monitoring is the regular habit of comparing what is happening against the plan, flagging the gaps, and acting on them. Data that is filed but never used has changed nothing. A monitoring system that never changes a decision is one that has failed at its real purpose, however neat its tables.

Reading · Types of evaluation

By timing

Baseline (start) · Midline (mid, for course correction) · Endline (close) · Ex-post (a year or two later, for sustainability).

By purpose

Formative (during, to improve the project) · Summative (at the end, to judge its overall worth).

Reading · Basic analysis, kept simple

Count

How many (e.g. 140 farmers adopted).

Percentage

Out of the whole (140 of 200 = 70%).

Disaggregate

Always break the total down by sex, age and location. An average can hide who was missed.

Compare

Endline against baseline (the change), and result against target (did you meet it).

Qualitative

Read through interviews and focus groups for recurring patterns and reasons.

Reading · From data to decision, and back to the community

A report is not the destination. The destination is a better decision. Hold short reflection moments: "given what this shows, what do we do differently?"

And account downward, not only upward. Take the findings back to the community that gave you the data — a feedback meeting, a durbar — in plain language, with a real way to respond. Downward accountability is not a courtesy. It is respect, and it makes your work better, because the community will often explain a finding you could not.