Impact Assessment

Impact assessments aim to determine the actual effects of an intervention, distinguishing between correlation and causation.

At Research360, we specialize in uncovering whether and how change occurred, and to what extent it can be attributed to the program in question.

We begin by clearly defining the counterfactual: what would have happened in the absence of the intervention. Depending on context, we employ experimental designs (like randomized controlled trials) or quasi-experimental designs (like difference-in-differences, propensity score matching, or regression discontinuity). These approaches allow us to isolate program impact from external influences.

Capturing long-term change requires tracking beneficiaries over time. We design and implement longitudinal studies that revisit the same individuals, households, or communities at multiple intervals post-intervention.

These studies provide powerful insights into sustainability, delayed effects, and life course transitions.

Our approach involves panel retention strategies, mobile follow-ups, and ethical protocols for sustained engagement. Longitudinal data enables better understanding of both immediate and evolving impact, providing donors with crucial evidence to inform scaling, redesign, or exit strategies.

Many real-world programs operate in complex environments where isolating impact is difficult. In these cases, we conduct contribution analysis to assess whether observed changes are plausibly linked to program activities.

This involves triangulating qualitative and quantitative data, examining stakeholder perceptions, and ruling out alternative explanations.

Contribution analysis doesn’t rely on counterfactuals but focuses on evidence coherence. This is particularly valuable in humanitarian, governance, or rights-based programs where change is collective and diffuse.

When feasible, we apply rigorous impact designs such as randomized controlled trials (RCTs), difference-in-differences, and propensity score matching to isolate program effects.

These methods help us separate causality from correlation and offer the highest level of confidence in attributing results.

When randomization isn’t possible, we use quasi-experimental methods with careful matching and baseline controls. These designs ensure reliability while remaining ethical and context-sensitive. We always balance statistical rigor with practical constraints to deliver credible and actionable results.

We use theories of change and logic models to frame how programs are expected to generate impact, helping identify not just what changed, but how and why. At the outset, we co-develop these frameworks with implementing partners to map inputs, activities, outputs, and anticipated outcomes, as well as assumptions and external risks.

Our evaluation tools are then aligned with these causal pathways to measure the results and validate the logic.