--- title: "Designing a Biological Experiment: Units, Controls and Randomisation" description: "Before any measurement is taken, the design fixes what can be concluded: what the experimental unit is, what varies with it, and what is held fixed or randomised. The design is written down before the" canonical: https://lightmysky.com/learn/science/designing-a-biological-experiment-units-controls-and-randomisation-mt_Hc6BAdlkLf source: https://lightmysky.com/learn/science/designing-a-biological-experiment-units-controls-and-randomisation-mt_Hc6BAdlkLf.md retrieved: 2026-11-12 --- > **Agent view.** This is the Markdown twin of the page, for tools and assistants. > When to use this site, and the call that answers each job: https://lightmysky.com/agent-instructions.md > API description (OpenAPI 3.1): https://lightmysky.com/openapi.json · Authentication: https://lightmysky.com/auth.md > Pricing: https://lightmysky.com/pricing.md · Catalog: https://lightmysky.com/llms.txt · Full catalog: https://lightmysky.com/llms-full.txt > Every machine-readable file on this domain: https://lightmysky.com/.well-known/ai-catalog.json > Ask for Markdown with `Accept: text/markdown`, a `.md` address, or `?mode=agent`. # Designing a Biological Experiment: Units, Controls and Randomisation Before any measurement is taken, the design fixes what can be concluded: what the experimental unit is, what varies with it, and what is held fixed or randomised. The design is written down before the data exist. Subject: Science · Area: Scientific Inquiry · Ages 22 to 23 Page: https://lightmysky.com/learn/science/designing-a-biological-experiment-units-controls-and-randomisation-mt_Hc6BAdlkLf ## Ready when they can - Identify the experimental unit in a stated protocol and count it correctly - Choose blocking or randomisation for a stated nuisance factor and justify the choice - Name a control that addresses a specified alternative explanation and state what it leaves open ## Lesson: Designing an experiment that can convince Before collecting study data, name the treatments, experimental units and response. Plan assignment, measurements and analysis. Random assignment balances other factors in expectation, not in every realised sample. Independent replication helps estimate uncertainty. Choose sample size for a stated effect and acceptable error rates. A causal conclusion still depends on how the study is carried out. An experimental unit is the smallest entity assigned independently to a treatment. It may be a patient, tray or cage. Several measurements within that unit do not become several treatment assignments. Count assigned units for treatment replication, and record how measurements are grouped within them. **Example.** Suppose four trays per treatment are assigned independently, with eighteen plants in each tray. There are four treatment units per arm. A tray mean can summarise its plants. The analysis compares independently assigned trays, rather than treating all plants as independent treatment replicates. *(drawing: A matched placebo compares the active drug with a similar pill without it. Other care should follow the same plan.)* Block on a known source of variation, such as clinic or laboratory batch. Group units by that factor, then randomise treatments within each block. This compares treatments within similar settings. Blocking does not replace random assignment. A clinic receiving only one treatment leaves clinic and treatment effects hard to separate. If a whole cage receives one treatment, the cage is the assigned unit. More animals in that cage add measurements, not independent cage assignments. With only one cage per treatment, cage differences cannot be separated from treatment differences. Plan several independently assigned cages per treatment to estimate between-cage variation. Random assignment supports treatment comparisons among study units. Random sampling addresses which population those units represent. A well-randomised volunteer study can still have limited generalisability. Controls address specified alternatives, such as effects of the vehicle or care schedule; no control removes every possible explanation. Human studies also require consent, risk review and privacy protections. **Recap.** Name assigned units, treatments and responses. Use suitable controls, randomise within planned blocks, and distinguish treatment replication from repeated measurements and population sampling. *Original lesson, questions and flashcards about biological experimental design, authored for this learning dataset.* ## Practice 8 questions on this page, each with its working shown. ## Needs first - [Errors, Power and the Design of a Test](https://lightmysky.com/learn/mathematics/errors-power-and-the-design-of-a-test-mt_D7ZXbvSD0l) - [Fractionation, Blotting and Pulling Down a Binding Partner](https://lightmysky.com/learn/science/fractionation-blotting-and-pulling-down-a-binding-partner-mt_MRfFQlBIPy) ## Opens up - [Pseudoreplication: When n Is Smaller Than It Looks](https://lightmysky.com/learn/science/pseudoreplication-when-n-is-smaller-than-it-looks-mt_2gRWB6RTaI)