Define the question before measuring
Time-series data can reveal whether a symptom follows weather, occupancy or equipment, but correlation still needs a plausible pathway.
Capture a useful baseline
Record the exact location, date, current weather or equipment state, recent activities and the first visible, audible, sensory or measurable change. Keep one unaffected reference point.
Safe observations to repeat
- Synchronize clocks
- Log trigger and symptom separately
- Keep sensor location fixed
Compare supporting and weakening evidence
Supports the pathway
- Events repeat with similar lag
- Control periods remain stable
- A mechanism explains the timing
Weakens the pathway
- Only one event exists
- Clock or sensor drift is present
- Many untracked variables change together
Interpret the result conservatively
Look for a repeatable relationship between trigger, location, timing and change. Treat exceptions as useful evidence rather than forcing every event into one explanation.
Escalation boundary
Do not infer causation from a coincidental graph without pathway evidence and competing explanations.
Close the loop
If work is performed, repeat the same observation under a comparable trigger and through the original lag window. Record both recurrence and non-recurrence events.
Primary references for this topic family
These references ground the general building-science or safety context. They do not diagnose a specific house, replace local requirements or certify a repair.
