1) Think about a question you are interested in answering
- A problem you have
- SOmething you want to know
- an experiment from your major you can replciate
- It must be measurable
- It must be definable
- It must be recommendable
- It must be interesting or useful
Example previous ideas
- What is the best place to see stars on campus (not OK - no reasonable way to count stars)
- What is the coolest place to sit outside on campus (not OK - no access to multiple thermometers and you need to measure it at the same time, and you have to define "in the trees")
- What is the quickest option for eating on campus (OK - measure times and recommend 1)
- What is the best way to get from this building to the ENgineering building (OK - measure times and recommend 1)
- Which uses most data - YouTube app or browser (OK - measure 2 data amounts and recommend 1)
2) Decide step by step how to measure it (in pairs, then unify plan in groups online)
Avoid problems with reliability and validity
- Definition problems
- Does "to the engineering building" mean the door or the path or a classroom?
- Measurement problems
- Problems with crowds interfering (do measurements in period 2 to avoid lunch crowds)
- Measurements were not equal (Before measuring set app and site resolution to same).
3) Read some literature
Usually this comes first, but no-one does experiments exactly like this so we're not reading for similar studies, we're reading for background.