The holistic CLD approach supports communities to collaboratively create a collective vision of the future they seek, and to work together to achieve that vision. On this page, you'll learn more about some key concepts of the holistic CLD approach and how they can be measured quantitatively. The Full Spectrum Evidence and Learning Initiative (FSELI) is a collaboration of NGOs, grassroots practitioners, and researchers committed to enabling a systemic shift in how holistic CLD is understood and implemented.
This tool presents some key concepts of the holistic CLD approach and how they can be measured quantitatively. Select a concept to learn more.
Many of the concepts presented here, or specific dimensions that emerged during the refinement process, were not captured in the existing quantitative monitoring instruments used by the FSELI implementing organisations. This measurement repository was developed to support evaluation design, focusing on guidance for measuring things that were not yet being measured quantitatively rather than documenting existing practice. Where overlap did exist, our measurement recommendations often diverged from the approaches taken in the original instruments, in part because some existing tools were designed with a qualitative lens that sits in tension with the core recommendations for quantitative measurement set out here.
A fuller description of Tostan and OVP's current monitoring efforts is also available in the FSELI blog: Building From Our Own Foundations: Using existing data to design a better evaluation. To hear more about the current measurement approaches of individual CLD organisations, reach out to Kyla Korvne at Tostan or Mohamed Rogers at One Village Partners.
Browse measurement guidance by theme. Choose a theme, then explore specific measurement dimensions.
A robust quantitative measurement strategy should prioritize capturing actual behaviors and outcomes, while ensuring that perceptions and attitudes are measured in a consistent and meaningful way. A few especially relevant principles for this work are:
When the objective is to capture behavior (e.g., participation, cooperation, conflict resolution), preference should be given to observed or revealed measures rather than self-reported data. These include:
Self-reported measures of behavior are often subject to recall errors and social desirability bias, and should be used with caution or triangulated with other sources.
For constructs that are inherently subjective, perception-based survey instruments are appropriate. These should rely on:
Where possible, measurement should draw on existing validated scales to ensure comparability and reliability, with that validation undertaken in comparable contexts.
A good measure produces comparable results whether collected by different enumerators, in different communities, or at different points in time. Investing in clear protocols, piloting, and enumerator training is essential especially in multi-site evaluations.
A measure is only useful if it can distinguish between communities or individuals that are genuinely different. Avoid scales or questions where most respondents cluster at one end, or where the instrument lacks the resolution to pick up meaningful variation.
More data is not always better. Lengthy surveys increase respondent fatigue, raise costs, and can reduce data quality. Prioritise a smaller set of well-chosen indicators over an exhaustive battery that covers everything loosely.
Background reading, methodological references, and validated measurement toolkits that inform this tool.
