A serious research schedule is not a list of weekly meetings. It is a sequence of decisions: narrow the question, confirm safety and feasibility, test the method, collect evidence, revise the interpretation, and prepare the student to defend the work independently. The calendar should protect that sequence rather than rush toward a poster date.
Weeks 1–4: define a question the student can own
The first month should create clarity, not a large dataset. In Dr. Li's mentorship model, the student compares prior work, narrows the problem, identifies a measurable outcome, inventories available resources, and explains why the proposed question is both useful and feasible.
- Build a small literature evidence table
- Write one testable question or engineering goal
- Define the primary variable, controls, and measurable output
- Identify equipment, local supervision, and likely constraints
Before experimentation: complete the plan and approval review
Society for Science requires a research plan before experimentation. Projects involving human participants, vertebrate animals, potentially hazardous biological agents, tissue, or certain hazardous activities may require formal review and approval before any data collection begins. This phase cannot be repaired retroactively by working faster later.
- Write materials, procedures, risks, and data analysis
- Separate the student's work from mentor or laboratory contributions
- Confirm the Adult Sponsor and direct supervision
- Obtain required approval before beginning restricted work
Pilot phase: test the measurement before scaling
A pilot asks whether the method produces a stable and interpretable signal. Depending on fabrication, access, and troubleshooting, this phase may take several iterations. A failed pilot is useful when the student records what failed, changes one decision at a time, and can explain why the revised method is better.
- Start with one baseline or control condition
- Repeat measurements to expose noise and drift
- Record failed tests and design revisions
- Stop and narrow the scope when the signal is unreliable
Evidence phase: collect only data the method can support
Once the measurement system is credible, the student can follow the approved procedure, preserve raw files, check quality during collection, and document deviations. More data do not rescue weak controls or a changing method, so milestone reviews should focus on evidence quality rather than sample count alone.
- Use the same documented procedure across trials
- Preserve dated notes, code, images, and original data
- Check controls and repeatability during collection
- Reapprove material plan changes when required
Analysis and defense: finish the reasoning, not just the experiment
The final phase should be reserved for figures, uncertainty, limitations, writing, and questions the student must answer without a script. Families should plan backward from the local affiliated fair and leave enough time for the student to connect every conclusion to traceable evidence.
- Analyze before designing the poster
- Distinguish observation from interpretation
- State limitations and alternative explanations
- Practice explaining each figure, decision, and contribution