EvalCompassEvalCompass
Practical evaluation guide

From evaluation findings to the next decision

A practical guide using fictional community-training results to establish denominators, missing data, interpretive limits, improvement decisions, responsibilities, deadlines and checking records.

Turn findings into action and follow-up

  1. What was checked

    Align the facts and denominators

  2. Interpretation

    Consider missing data and differences

  3. Next decision

    What will change, and what will be checked?

  4. Responsibility and deadline

    Who will act, and when?

  5. Follow-up

    Check results and share them back

From the beginning, plan what will happen after the evaluation report is submitted. Read the full explanation

Main takeaway

Create records linking confirmed findings → interpretation → next decision → responsibility and deadline → review. Plan actions after report submission from the start of evaluation. [1]

The community watch training here is fictional teaching material. Numbers, proportions, schedules and decision conditions are illustrative settings, not standard effects or durations.

One example connects training offered to 20 applicants to records after 4 weeks and decisions for improving the next session.

Decide the next actions from evaluation findings

Distinguish facts from interpretations, then connect judgments, responsibility, and reporting back to the people involved.

1. Align facts and denominators

All organizations, numbers, proportions, dates and judgments here are fictional. In the preceding guide's watch training, suppose there were 20 applicants, 18 attendees, 15 responses after 4 weeks and 10 reports of procedure use. Label each number with whom it concerns.

This does not show that training produced a 66.7% behavior implementation rate. There is no comparison condition; other learning, prior experience, opportunities and non-response also matter.

RecordCalculationMeaning
Training attendance18 / 20 = 90%Attendance rate among applicants; output reach.
Follow-up responses15 / 20 = 75%Response rate among applicants; behavior of 5 remains unconfirmed.
Respondents' behavior implementation10 / 15 ≈ 66.7%Self-report; assume the remaining 5 responses were 3 without opportunities and 2 non-users.
Confirmed implementation among all applicants10 / 20 = 50%Self-reported implementation confirmed for half of applicants; this does not establish the true overall implementation rate as 50%.
Practical framework or suggestion
[2]

2. Interpret missing data and differences

If all 5 non-respondents did not use procedures, the overall proportion is 10/20=50%; if all did, it is 15/20=75%. The 50–75% range uses extreme assumptions about these 5 missing responses only. It is not a confidence interval, nor an uncertainty range covering self-report errors or target-group representativeness.

Averages can conceal people facing participation difficulties, so check necessary subgroups. In this example, non-response is concentrated among experienced participants. Fine-grained characteristics may identify people; the table below is for teaching only.

The experienced group's 4/5 does not establish stronger training effects for experienced participants, because non-respondent status is unknown and groups are small. Interviews about difficulties help seek missing voices and contrary cases. Do not convert interview reasons into proportions of all 20.

CategoryApplicantsResponsesSelf-reported useNo response
First-time participants10 people10 people6/10 people0 people
Experienced participants10 people5 people4/5 people5 people
Total20 people15 people10/15 people5 people
Practical framework or suggestion
[3]

3. Decide actions and the next experiment

Match the strength of conclusions to the scale of decisions. A small reversible teaching-material trial and large expansion elsewhere require different evidence.

When using DAC criteria, connect questions to relevance of accessibility and efficiency of consultation as well as effectiveness. Record reasons for adopted perspectives and those deferred, rather than mechanically scoring 6 items.

CategoryWhat to record
Confirmed factsOf 15 respondents, 10 used procedures, 3 had no opportunity and 2 did not use them. Interviews included difficulty imagining situations for using procedures.
Interpretation and hypothesisSituation-based exercises and opportunities for post-participation consultation may support use.
UnconfirmedStatus of 5 non-respondents, consistency between self-report and actual procedures, whether training caused use, and who can access consultation.
Current decisionTry situation-based exercises and voluntary consultation on a small scale next time. Improve materials and recheck reasons for use, lack of opportunities and difficulties.
Next design for testing effectsTo establish additional exercise effects, separately design next-session comparisons, assignment and main indicators before implementation. Counts before and after improvement alone do not demonstrate effects.
Practical framework or suggestion
[4] [5]

4. Set responsibilities, deadlines and checking records

Make actions in response to findings checkable in routine operational meetings. Go beyond improve and record who changes what and how it will be checked.

Check implementation separately from whether change occurred. Record reasons for plan changes when unexpected changes, local circumstances or practice opportunities shift.

What to changeResponsibility / deadlineRecords to checkConditions for review
Add 1 situation-based exerciseTraining lead / December 17Material revision history and trial participants' understanding and difficultiesRevise if time burden or misunderstanding increases.
Offer choices for follow-up contactContact lead / before next recruitmentReach, response counts and target numbers by contact methodDo not force responses; change if burden concentrates on particular people.
Try voluntary consultation opportunitiesFrontline staff / 4 weeks after next trainingUse, difficulties and staff time burdenReview information or hours if some still find access difficult.
Report findings and changes back to affected peopleOperations lead / 1 week after the meetingBrief explanation, delivery methods and differing feedback receivedCorrect or protect if factual errors or publication concerns arise.
Practical framework or suggestion
[1] [6]

5. Report back and reconsider value

Briefly explain what is known, what remains unknown, what will change and how to give feedback. Check identification through small-group characteristics or detailed accounts, and respect the consent scope for responses and publication.

When addressing social value, center changes and improvements important to affected people. Under SVI2026 organizational social value reporting standards, SROI ratio disclosure in accounting notes is optional; disclosure requires assumptions, sensitivity analysis and scope. Do not make a single ratio the sole completion criterion.

Fictional feedback: 18 people attended training. Of 15 responses after 4 weeks, 10 reported using procedures. The status of 5 non-respondents and the effect of training itself remain unconfirmed. Next time we will try situation-based exercises and voluntary consultation and check barriers to use. You may voluntarily send feedback to the contact lead.

Bring known findings, unconfirmed matters, changes, responsibilities and deadlines, and checking records to the next meeting.

Practical framework or suggestion
[7] [8]
Fictional example; not actual results

Connect applicant follow-up findings to decisions

A fictional example for illustration: 20 applicants, 18 attendees, 15 follow-up responses and 10 self-reports of procedure use.

Results to checkScope of interpretationNext decision
10/15 ≈ 66.7%Respondents' self-reported use; 5 non-respondents unconfirmedState denominators and voluntarily check non-response concentrated among experienced participants
10/20 = 50%Confirmed implementation among all applicants; not an established true overall proportionExplain the 50–75% range under extreme missing-data-only assumptions, distinct from a confidence interval
3 without opportunities; 2 non-usersDo not equate opportunities with lack of understanding or supportPlan small trials of situation-based exercises and voluntary consultation with owners and deadlines

What this example supports

Distinguish observed change from causal training effects. Do not decide expansion from ratios alone; retain unknowns, burden, affected people's voices and next checks.

Common errors in judgment

Treat 10/15 as 66.7% of all 20

How to revise: Report respondents' 66.7%, confirmed use among all applicants of 50%, and 5 non-respondents together

Reason: Denominators and missing data change meaning.

Present the extreme-assumption 50–75% range as a 95% confidence interval

How to revise: Explain that it assumes all 5 non-respondents used or did not use procedures, and is not a confidence interval

Reason: It does not estimate sampling uncertainty or include other uncertainty such as self-report error.

Put it into practice next

Record users, decision timing, findings, actions, responsibilities and checks in the evaluation use worksheet

View an example and open the worksheetRead the method explanation

Evaluation methods to explore next

Evaluation frameworksUtilization-Focused Evaluation
An approach that clarifies who will use evaluation, for which decisions and w…
Data collectionInterview
A method of eliciting specific experiences and reasons for judgments through …
Design & theoryEvaluation Design Matrix
A table linking evaluation questions, judgment criteria, required information…
Design & theoryIndicator Design
Specify the desired change through definitions, sources, collection methods a…

Frequently asked questions

Does before-and-after comparison show program effects?

It can establish observed changes, but other factors also contribute, so before-and-after differences alone do not establish causal effects. Design comparison conditions and methodological assumptions separately.

Can non-respondents be excluded from aggregation?

Respondent aggregation is possible, but state the denominator, total targets, non-respondent numbers and bias. Do not present respondent-only proportions as overall proportions.

References and checked sections

  1. Report & support use of findings
    BetterEvaluation / Global Evaluation Initiative
    Section consulted
    opening paragraphs / Identify reporting requirements / Ensure accessibility / Develop recommendations / Support use
    How it relates to the text
    Identify users, decision timing and reporting formats from the start and support action on findings.
    Source checked: (includes AI cross-checking)
  2. CDC Program Evaluation Framework, 2024
    Centers for Disease Control and Prevention (CDC), MMWR
    Section consulted
    Types of Evaluation / Step 1: evaluability assessment / Step 2: program roadmap / Step 4: Indicators and data sources / Step 6: Act on findings
    How it relates to the text
    6 stages from context to action, with collaboration, equity, learning and use across stages. Indicators can concern activities or outcomes; outcome checks alone do not establish attribution.
    Source checked: (includes AI cross-checking)
  3. Quality in qualitative evaluation (QQE)
    HM Treasury / Evaluation Task Force
    Section consulted
    §4.2: guiding principles / Framework questions 4, 7–8, 12: inference, sample coverage, diversity
    How it relates to the text
    Qualitative evaluation claims require analytical processes, case selection, different voices, and evidence and limits for broader inference.
    Source checked: (includes AI cross-checking)
  4. Magenta Book: Central Government guidance on evaluation
    HM Treasury / Evaluation Task Force
    Section consulted
    §1.6 pp.23–24: early planning / §2.2: evaluation scoping / §3.2–3.5 pp.71–84: method choice and causal approaches / §6.2 pp.144–145: use and dissemination
    How it relates to the text
    Integrating evaluation early into program design and selecting methods suited to questions and use. Impact evaluation distinguishes observed change from causal contribution.
    Source checked: (includes AI cross-checking)
  5. Applying Evaluation Criteria Thoughtfully
    OECD Publishing
    Section consulted
    Executive summary: two principles / Chapter 3: Applying criteria thoughtfully / Chapter 3: Choosing which criteria to use
    How it relates to the text
    The 6 criteria are perspectives for value judgments, selected and applied carefully according to purpose, context and stakeholder power relations.
    Source checked: (includes AI cross-checking)
  6. Supplementary Guide: Handling complexity in policy evaluation
    HM Treasury / Evaluation Task Force; developed by CECAN
    Section consulted
    Chapter 1: context/history, continuous learning / Chapter 2: Table 2 evaluation challenges / Chapter 4: complexity-appropriate approaches
    How it relates to the text
    Track context, interactions and unexpected change in complex programs and revise designs as learning develops.
    Source checked: (includes AI cross-checking)
  7. UNEG Ethical Guidelines for Evaluation
    United Nations Evaluation Group (UNEG); official copy hosted by UNODC
    Section consulted
    pp.7–10: integrity, accountability, respect, beneficence / pp.20–22: conduct, consent, confidentiality, power / Annex 3 pp.30–35: phase checklists
    How it relates to the text
    Applying integrity, accountability, respect and beneficence throughout; specifying consent, withdrawal, confidentiality scope and harm reduction.
    Source checked: (includes AI cross-checking)
  8. Standards for Applying the Principles of Social Value: The Complete Set
    Social Value International
    Section consulted
    §4.4.3 p.21: sensitivity of prioritisation / §5.1–5.3 p.23: counterfactual, attribution, displacement / Section Three p.33 items 8–9: SROI ratio optional, sensitivity, recommendations / Appendix 3 p.37: ratio as range and assumptions
    How it relates to the text
    Social value analysis concerns changes for affected people and decisions. SROI ratios are optional in the 2026 reporting items; disclosure includes sensitivity, assumptions and scope.
    Source checked: (includes AI cross-checking)
Content updated: Source checked: (includes AI cross-checking)No documented human reviewIncludes fictional examplesEditorial policy and scope of checkingCite this pageReport a correction

How to cite this page

EvalCompass Editorial Team. “From evaluation findings to the next decision”. EvalCompass, updated 2026-10-07. https://evalcompass.jun-nakatani.com/en/guides/use-findings/
Add the date you accessed the page. Text created by this site is available under CC BY 4.0. Quotations, external materials, and bundled fonts have their own terms of use. When reusing content, credit the source, link to the license, and indicate any changes.

Suggest a correction or improvement to this page

Change history
  • — A cross-guide review clarified indicator formats, outcome classifications and causal judgments, interpretation of tables, and the scope of SVI2026 where relevant.
  • — Added a complete fictional community-training example linked to official primary sources. Specified denominators, missing data, causal scope and responsibility and timing for improvement decisions.