I85PREP
⌘ K
CNP study materials
CNP study resources

CNP: Domain 4: Program Planning, Evaluation, and Impact

CNP · Work through your exam, one topic at a time.

All guide sections

Domain 4: Program Planning, Evaluation, and Impact

What the exam tests

Program Planning, Evaluation, and Impact tests logic models, needs assessment, SMART objectives, process and outcome evaluation, and how to report credible impact.

Core rules

  • Logic model: inputs → activities → outputs → short-term outcomes → intermediate outcomes → impact. Outputs count activity (sessions delivered, people served); outcomes describe change in participants, organizations, or communities.
  • Theory of change explains the causal assumptions behind the model-why these activities should produce those outcomes.
  • SMART objectives: specific, measurable, achievable, relevant, and time-bound. "Increase daily attendance among participants by 15% by June 30" is SMART; "help youth succeed" is not.
  • Needs assessment: use community data, key informant interviews, surveys, and asset mapping before designing a program; include intended participants in the design.
  • Evaluation types: formative (improve during implementation), summative (judge results), process (was it delivered as intended), outcome (did participants change), and impact (longer-term; attribution is harder).
  • Indicators need baselines, targets, data sources, and a collection schedule. Use both quantitative and qualitative data.

Worked example

An after-school program reports: 120 students attended at least 20 sessions; 78% improved school attendance by 10% or more; 65% of participants' teachers reported better classroom engagement. The first number is an output; the second and third are outcomes. A funder asking "What impact did you have?" should get the outcome data plus a caution: without a comparison group, the program can claim contribution, not sole causation.

Traps that cost marks

  • Presenting outputs as outcomes ("we served 500 people" is not impact).
  • Having no baseline, so change cannot be measured.
  • Evaluating only at the end, with no feedback loop for improvement.
  • Claiming attribution without a comparison group or credible design; say "contributed to" rather than "caused."
  • Ignoring unintended consequences and participant voice.