Design a Sampling Plan
Design a sampling plan with frame, method, size, and selection criteria.
Draft: Design a Sampling Plan
Fill In Your Details (4 Blanks)
Add your information to get a tailored prompt. You can also use it as is.
Sector and Donor Context (Optional)
Your Prompt (With Your Details)
This is the complete prompt, ready to copy and paste.
You are a senior MEAL specialist designing a sampling plan for your program to support a planned study (e.g., baseline survey, midline assessment).
The plan must be explicit about target population, frame, method, size, and selection, with the trade-offs stated honestly.
Requirements:
- Target population. Define the target population precisely, including the relevant inclusion criteria.
- Sampling frame. Identify the frame (e.g., enrollment registry, household roster, geographic listing). State its known gaps and biases.
- Sampling method. Choose a method (e.g., simple random, stratified random, multi-stage cluster, purposive). Justify the choice in relation to the study's questions and feasibility.
- Sample size. Calculate or propose a sample size with the underlying assumptions (e.g., confidence level, margin of error, design effect, expected response rate). For purposive samples, justify saturation logic.
- Stratification and disaggregation. Specify stratification variables and ensure each stratum has adequate cases for planned disaggregated analysis.
- Selection and recruitment. Describe how units will be selected and approached, including replacement rules and refusal handling.
- Limitations. State known limitations honestly (frame coverage, non-response bias, generalizability boundaries).
Output Format:
Produce:
- A 1-paragraph framing.
- A sampling plan table (component, decision, justification).
- A sample size calculation with assumptions shown.
- A short note on how the plan will be quality-assured during fieldwork.
- A limitations paragraph.
Check: Score It with the Sampling Design Quality Rubric
The Review Prompt
Paste the draft at the bottom of this prompt and run it. It scores each criterion from 1 to 5 and says why.
You are an expert M&E sampling specialist. Score the sampling design I will provide using the rubric below.
SCORING RUBRIC - Sampling Design Quality Score each dimension 1-5 using these criteria:
DIMENSION 1: Sampling Strategy and Justification
- Score 5: All four elements present and well-reasoned. Sampling approach (probability, purposive, mixed) is named, rationale articulates why it was chosen over alternatives, rationale connects explicitly to evaluation questions, and the choice is appropriate for the stated inference goals (generalization, depth, comparison).
- Score 4: At least three of four elements present. Approach named with rationale. Rationale may be brief or weakly connected to evaluation questions.
- Score 3: At least two of four elements present. Approach named but rationale generic. Connection to evaluation questions implicit.
- Score 2: One element present (typically just naming the approach). No rationale or evaluation-question connection.
- Score 1: No sampling strategy stated, OR strategy clearly mismatched to inference goals.
DIMENSION 2: Sample Size and Power
- Score 5: All four elements present. Sample size stated, calculation method shown (statistical power calculation for quant, saturation or information-power criteria for qual), assumptions documented (effect size, design effect, response rate, expected variability), non-response buffer included.
- Score 4: At least three of four elements present. Size and calculation method present; assumptions or buffer partially documented.
- Score 3: Size stated and calculation method named, but assumptions or buffer not documented. For qualitative: size stated but saturation criteria weakly defined.
- Score 2: Size stated but no calculation or rationale. Quantitative size cannot be tied to power; qualitative size cannot be tied to saturation.
- Score 1: No sample size stated, OR size grossly inadequate for stated aims.
DIMENSION 3: Selection Method
- Score 5: All four elements present. Sampling frame named and its quality assessed, selection mechanism specified (random, systematic, stratified, purposive criteria), stratification or clustering rationale given where used, replacement and non-response rules stated.
- Score 4: At least three of four elements present. Sampling frame and mechanism named; stratification logic or replacement rules incomplete.
- Score 3: At least two of four elements present. Sampling frame and selection mechanism named, but stratification logic or replacement rules missing.
- Score 2: Selection mechanism named but sampling frame absent or unsuitable, OR mechanism is undefined ("sample 100 households" without specifying how).
- Score 1: No selection method described.
DIMENSION 4: Representativeness and Coverage
- Score 5: All four elements present. Target population precisely defined (inclusion and exclusion criteria), exclusions justified, geographic and demographic coverage mapped, GESI representation deliberate (sex, age, disability, other relevant identities).
- Score 4: At least three of four elements present. Target population defined and coverage mapped; GESI representation or exclusion rationale partial.
- Score 3: Target population defined, but coverage or GESI representation generic. Exclusions not justified.
- Score 2: Target population vague or contradicts the inference goals. Coverage gaps significant.
- Score 1: No defined target population, OR sample obviously biased away from the population of interest.
DIMENSION 5: Documentation and Replicability
- Score 5: All four elements present. Written sampling plan exists (separate document or clear methodology section), log of selected/rejected/replaced units maintained, weights or post-stratification specified for probability samples with unequal selection probabilities, audit trail of decisions made during fielding.
- Score 4: At least three of four elements present. Sampling plan and log exist; weights or audit trail incomplete.
- Score 3: At least two of four elements present. Sampling plan exists. Log informal or partial. Weights not specified or not applicable. No audit trail.
- Score 2: Sampling plan minimal or scattered. No log. Replicability would require substantial reconstruction.
- Score 1: No documentation. Sampling cannot be replicated by another researcher.
OUTPUT FORMAT: Return your assessment as a table followed by a summary:
| Dimension | Score (1-5) | Evidence from Sampling Design | Priority Revision |
|---|---|---|---|
| Sampling Strategy and Justification | |||
| Sample Size and Power | |||
| Selection Method | |||
| Representativeness and Coverage | |||
| Documentation and Replicability |
Total: X/25 Band: Strong (22-25) / Adequate (17-21) / Needs Revision (11-16) / Substantial Revision (5-10) Single Most Important Revision: [One specific sentence]
For any dimension scored 1 or 2, add a brief explanation and a concrete revision example.
SAMPLING DESIGN TO SCORE: [Paste your sampling design here]
Fix: Revise What Scored Low
The Revise Prompt
Paste the draft and its scores at the bottom of this prompt. It revises the parts that scored below 4.
You are an expert M&E sampling specialist. Review the draft sampling design I will provide and fix every problem that falls below the Good (4) threshold on the Sampling Design Quality rubric.
WHAT TO LOOK FOR AND FIX:
Sampling Strategy problems: Approach not named. No rationale or generic rationale. No connection to evaluation questions. Mismatch between approach and inference goals (e.g., purposive sampling claimed to support population-level generalization).
Sample Size problems: Size stated without calculation. Missing assumptions (effect size, design effect, response rate). No non-response buffer. Qualitative size without saturation or information-power criteria.
Selection Method problems: Sampling frame not named, or named but unsuitable for the target population. Selection mechanism vague ("sample 100 households" without method). Stratification used without rationale. No replacement or non-response rules.
Representativeness problems: Target population vague or undefined. Exclusions unjustified. Significant geographic or demographic coverage gaps. No GESI representation plan.
Documentation problems: No written sampling plan. No rejection/replacement log. Weights not specified for unequal probability samples. No audit trail.
OUTPUT FORMAT: For each problem found:
- Dimension: [which dimension]
- Location: [which section of the sampling design]
- Problem: [one sentence describing what is wrong]
- Revised version: [the corrected text]
Then provide the complete revised sampling design incorporating all corrections.
DRAFT SAMPLING DESIGN TO REVISE: [Paste your draft sampling design here]