Sampling Plans and Representative Measurements
Design sampling that connects the stated population and decision to representative spatial, temporal, and biological observations.
Design sampling that connects the stated population and decision to representative spatial, temporal, and biological observations.
Core science
A measurement describes the material actually observed or submitted. Extending it to a plant, room, irrigation zone, field, batch, or cultivar requires a sampling design. Begin by defining the target population and decision. A “room average” can refer to sensors, plants, canopy positions, or time points. A “batch sample” can refer to a few selected flowers, randomized increments, a composite, or material chosen for appearance. These designs support different inferences.
Heterogeneity should guide stratification. Useful strata include cultivar, propagation lot, room zone, canopy height, irrigation line position, soil unit, flowering stage, drying rack, container depth, and time. Random selection reduces conscious preference, while systematic routes improve repeatability but can align with hidden patterns. Composite samples estimate an average but can conceal hotspots; individual increments preserve spatial information but cost more to analyze.
Sample integrity continues after collection. Tools, containers, labeling, temperature, moisture loss, transport time, grinding, homogenization, subsampling, and chain of custody can alter results. Replicate laboratory analyses do not repair a biased field sample. A sampling plan should state how missing units, damaged samples, outliers, and deviations will be handled before results are known.
Why this matters in cultivation
- Write the sampling plan before inspection or testing. Include a map, selection rule, number of units, sample mass or volume, handling, retained samples, and inference limits.
Measure and record
Record 1
Define the target population, decision, sampling frame, sampling unit, increment or composite rules, strata, and any inaccessible portions before collecting material.
Record 2
Record the randomization or route, number of samples and rationale, exact locations and times, sample mass or volume, tools, containers, preservation, chain of custody, and rules for missing or unusable samples.
Record 3
At interpretation, state what population the sample is intended to represent, how compositing may hide spatial variation, whether retained samples exist, and which rare or localized conditions the design could reasonably miss.
Common misconceptions
Evidence limits and uncertainty
Representative sampling reduces sampling error but cannot eliminate uncertainty from rare hotspots, inaccessible units, temporal change, destructive testing, or incomplete frames.
Analytical sensitivity cannot compensate for poor population coverage; a highly accurate result from a biased sample remains a biased estimate of the batch or crop.
Check your reasoning
- For "Sampling Plans and Representative Measurements", which records are required to make the result traceable and decision-ready, and which missing field would most weaken the conclusion?
- A learner claims, "A larger composite always detects localized contamination." Use the lesson’s science and evidence limits to explain why that claim is unreliable, then name one observation or measurement that could separate the competing explanations.
- Applied case — Write the sampling plan before inspection or testing. Include a map, selection rule, number of units, sample mass or volume, handling, retained samples, and inference limits. Build a verification plan using the lesson’s record set (Target population and decision; sampling frame; unit/increment/composite definition; strata; randomization or route; sample count and rationale; locations/times; sample mass/volume; tools/containers; preservation; chain of custody; missing/deviation rule; retained sample; inference statement.). What would you compare before and after the action, and what result would make you revise the original interpretation?
Require lesson-specific evidence, not memorized universal targets. Open the rationales after you have written or discussed your own answer.
Answer rationale 1: Mechanism / workflow rationale
- A strong answer should connect the response to the lesson objective: Design sampling that connects the stated population and decision to representative spatial, temporal, and biological observations.
- A measurement describes the material actually observed or submitted. Extending it to a plant, room, irrigation zone, field, batch, or cultivar requires a sampling design. Begin by defining the target population and decision. A “room average” can refer to sensors, plants, canopy positions, or time points. A “batch sample” can refer to a few selected flowers, randomized increments, a composite, or material chosen for appearance. These designs support different inferences.
- Heterogeneity should guide stratification. Useful strata include cultivar, propagation lot, room zone, canopy height, irrigation line position, soil unit, flowering stage, drying rack, container depth, and time. Random selection reduces conscious preference, while systematic routes improve repeatability but can align with hidden patterns. Composite samples estimate an average but can conceal hotspots; individual increments preserve spatial information but cost more to analyze.
- The most useful verification evidence includes Define the target population, decision, sampling frame, sampling unit, increment or composite rules, strata, and any inaccessible portions before collecting material..
- Keep this limit explicit: Representative sampling reduces sampling error but cannot eliminate uncertainty from rare hotspots, inaccessible units, temporal change, destructive testing, or incomplete frames.
Answer rationale 2: Misconception rationale
- The shortcut is unreliable because the lesson explicitly teaches a more conditional explanation.
- Representative misconception: A larger composite always detects localized contamination. Compositing can dilute a hotspot below detection even while improving estimation of an average condition.
- A measurement describes the material actually observed or submitted. Extending it to a plant, room, irrigation zone, field, batch, or cultivar requires a sampling design. Begin by defining the target population and decision. A “room average” can refer to sensors, plants, canopy positions, or time points. A “batch sample” can refer to a few selected flowers, randomized increments, a composite, or material chosen for appearance. These designs support different inferences.
- A useful discriminator is Record the randomization or route, number of samples and rationale, exact locations and times, sample mass or volume, tools, containers, preservation, chain of custody, and rules for missing or unusable samples..
- Do not overextend the conclusion beyond this limit: Representative sampling reduces sampling error but cannot eliminate uncertainty from rare hotspots, inaccessible units, temporal change, destructive testing, or incomplete frames.
Answer rationale 3: Applied verification rationale
- In practice: Write the sampling plan before inspection or testing. Include a map, selection rule, number of units, sample mass or volume, handling, retained samples, and inference limits.
- Record before action: Define the target population, decision, sampling frame, sampling unit, increment or composite rules, strata, and any inaccessible portions before collecting material..
- Also record: Record the randomization or route, number of samples and rationale, exact locations and times, sample mass or volume, tools, containers, preservation, chain of custody, and rules for missing or unusable samples..
- After the action, repeat the same measurement or observation so the comparison is valid.
- Revise the interpretation if the result conflicts with the lesson limit or the expected response: Representative sampling reduces sampling error but cannot eliminate uncertainty from rare hotspots, inaccessible units, temporal change, destructive testing, or incomplete frames.
Worked example: One sensor is used to represent an entire room
Scenario: A cultivation room uses one centrally located sensor to support a claim that all plants experienced the same temperature and RH.
Reasoning path
- Define the population and decision: whether the claim concerns the whole room, canopy zones, time periods, or a specific crop operation.
- Identify likely spatial and temporal gradients created by fixtures, walls, doors, airflow, irrigation, and canopy density.
- Choose representative sampling locations and times before collecting the data.
- Distinguish repeated readings from one sensor from independent spatial samples.
- Use the resulting variation to decide whether one fixed sensor is representative enough for the intended decision.
Evidence to collect
- room/canopy map
- sensor positions
- time intervals
- equipment/airflow layout
- independent spatial readings
- sensor verification status
- decision threshold or tolerance
Common weak answers
- A central sensor automatically represents the room.
- Many readings from one location equal many independent samples.
- An average proves there were no harmful extremes.
Verification: A representative plan should reproduce the important spatial and temporal variation well enough to support the stated decision, with limits documented.
Applicability boundary: Sampling density depends on room geometry, variability, equipment, risk, and the consequence of missing extremes.
Related lessons
Sources and evidence
- ISO/IEC 17025:2017 — General requirements for the competence of testing and calibration laboratoriesV21-SRC-008
Laboratory competence, impartiality, method control, equipment, traceability, sampling, reporting, and nonconforming work; standard text and current confirmation status require controlled access.
- NIST/SEMATECH e-Handbook of Statistical MethodsV21-SRC-009
Primary reference for experimental design, measurement process characterization, control charts, uncertainty, regression, and statistical graphics.
- JCGM 106 — The role of measurement uncertainty in conformity assessmentV21-SRC-019
Decision rules, guard bands, acceptance limits, and uncertainty in pass/fail decisions.
- ASTM Committee D37 on CannabisV21-SRC-020
Current consensus standards and work items for cannabis sampling, testing, quality, security, and processing; exact edition and applicability must be verified.
- EPA — Quality System and Quality Assurance Project PlansV21-SRC-026
Data-quality objectives, sampling design, QA/QC, documentation, validation, and corrective action.
Downloads
No lesson-specific download is approved for this release. Use browser print/save-to-PDF when you need an offline reading copy.