What is a baseline survey? Endline differences, design and cautions

A baseline measures starting conditions; an endline measures conditions at program completion or after a period. Consistent targets, definitions and similar conditions make differences easier to interpret.

For employment support, choose employment status, application behavior or other items with consistent definitions before and at completion. Plan a separate follow-up point if continuity after completion matters.

First select questions and main changes to examine at completion. Rather than fixing item counts, check answerability, consistent definitions and sustainable recording.

Generally align baseline and endline question wording, options, scales and aggregation units. If changes are necessary, record them and explicitly report that simple comparison may not be possible.

1. Decide evaluation questions and main outcomes
2. Inventory data available before starting
3. Minimize additional indicators and questions
4. Decide targets, timing, methods and responsibilities
5. Check that identical definitions can be used at completion
6. Record missing data, attrition and external factors

Numbers with identical formats may measure different meanings. Changes in wording, question order or face-to-face, phone and online modes may affect responses. Where changes are necessary, check their effects and record comparison limitations.

Turnover in targets, attrition, seasonality and institutional changes also affect results. When change appears, consider explanations other than the program and supplement context through interviews or records as needed.

• Align indicator definitions, units and response scales
• Decide whether to follow the same individuals or compare whole groups
• Retain initial and final response rates and missing-data reasons
• Record major external factors and implementation-condition changes
• Interpret quantitative differences with qualitative information

Before-and-after change need not arise from the program alone. Natural changes, other support or social conditions may influence results. Before-and-after comparison helps establish outcomes, but report it separately from demonstration of causal effects.

For stronger causal examination, consider comparison groups, multiple pre-intervention timepoints or contribution analysis. For small programs, practical decisions with explicit comparison limitations may be realistic.

• Do not confuse before-and-after change with causal program effects
• Check comparison groups and opportunities to examine alternatives
• Distinguish individual changes from overall averages
• Include data limitations in conclusions