С чего начать · Начальный
Template 1: Research question and project proposal
Use this worksheet before committing to data collection. Replace bracketed prompts with specific information. Keep unresolved points visible until they are resolved.
3 мин чтения · 26 окт. 2026 г.
Эта страница ещё не переведена, поэтому вы читаете английскую версию.
Главное
- Adapt the prompts to the actual study.
- Complete every statement from verified records.
- Follow the target journal and institutional requirements.
Содержание
Use this worksheet before committing to data collection. Replace bracketed prompts with specific information. Keep unresolved points visible until they are resolved.
Use this worksheet before committing to data collection. Replace bracketed prompts with specific information. Keep unresolved points visible until they are resolved.
Project identity
Working title: [Specific subject and setting]
Researcher and supervisor: [Names and roles]
Version and date: [Version; date]
Question and rationale
Research question: In [population or system], how does [condition, exposure or process] relate to [defined outcome or experience] in [setting and period]?
Question type: [Descriptive, association, causal, prediction, methodological or qualitative]
What is already known: [Two or three supported statements with verified sources]
What remains uncertain: [A precise limitation of the available evidence]
Why the answer matters: [What understanding or decision could change]
Contribution: [Replication, new evidence, method comparison, synthesis or another defined contribution]
Aim and evidence
Primary objective: We will [estimate, compare, describe or interpret] [specific quantity or phenomenon].
Primary estimand, where applicable: [Population, outcome, comparison and summary quantity]
Hypothesis, if appropriate: [Testable expectation and its rationale]
Exploratory questions: [Clearly separate later or broader questions]
Operational definitions: [Define terms such as improvement, success, confidence or accuracy]
Evidence that would answer the question: [Observations and analysis required]
Evidence that would leave it unresolved: [Examples of inadequate precision, missing comparison or alternative explanation]
Feasibility and responsibility
Data or materials: [Source, access and licence]
Design: [Brief outline and independent unit]
Required expertise and resources: [People, equipment, software and time]
Ethics or permission route: [Responsible institution or body and current status]
Main risk to a useful result: [Specific scientific or practical limitation]
Smallest useful project: [A reduced scope that still answers a meaningful question]
Next decision and date: [Action needed before proceeding]
Worked starter
For the synthetic robotics teaching example: “Estimate the mean paired difference in positioning error between two controllers across a defined class of layouts.” A real proposal must additionally define how layouts are sampled and how error is measured. The teaching dataset alone does not provide that design.
Tutorial: turn the proposal into a decision
Read Article 1 before completing this sheet. Begin with the decision your evidence could inform, then choose the outcome and the design. Do not start by naming a statistical test. Write a second sentence stating what the project cannot establish. This sentence helps prevent a narrow measurement from becoming a broad conclusion later.
Hypothetical worked choice: A student interested in urban plants proposes to compare leaf surface dust at two roadside locations. The independent sampling unit might be a plant, with several leaves measured within each plant. A defensible question concerns a difference between those locations during the sampling period. A claim that traffic causes the difference would require stronger evidence about competing environmental explanations. The proposal should therefore record location selection, plant selection, species, measurement timing and the intended boundary of interpretation.
Decision to document: Explain why this comparison matters even if no clear difference appears. A feasibility study, a replication or a carefully bounded descriptive study can have a useful purpose. Describe the contribution in terms of information gained, rather than the hoped for direction of the result.
Completion check: Give the proposal to someone who has not discussed the project with you. Ask them to identify the question, the independent unit, the outcome and the strongest permissible conclusion. Revise any part they cannot identify from the written plan.