AtlasSession 4L10: Sample size calculationsp.6–7
Continuous outcomes
- 7 earlier topics
- Tests, p-values, significance
- CIs and p-values
- Beta error and power
- Why sample size matters
- Continuous outcomes
- Next: Minimal important difference
Fig. 1
Difference in mean compressive modulus, crosslinked minus standard hydrogel (MPa)
Fig. 1. Two overlapping humps: the results you would get if there is no real difference, and if there is a real difference of 0.50 MPa. With 30 hydrogels per group, about 49 in 100 studies would spot the real difference and about 51 in 100 would miss it. Spotting it about 80 in 100 needs 63 hydrogels per group.
- Studies that spot it
- about 49 in 100
- Studies that miss it
- about 51 in 100
- hydrogels per group
- 30
- Needed to spot it about 80 in 100
- 63 hydrogels per group
Two-Curve Test Visualiser · First look
Level 1 · First look · Lecture 10 · p.6–7
Say in plain words what makes a study need more samples: a smaller difference, more scatter, or more certainty.
List the four inputs for a continuous-outcome sample size (α, β or power, difference in means, SD of both groups), say where each comes from, and predict how n moves when each changes.
Spotting a difference (the step you are on)
Two bakeries' loaves; a big weight gap shows with few loaves, a tiny one needs many.
Like loaves from two bakeries, a big gap in stiffness shows with few samples: at 30 hydrogels per group, about 49 in 100 repeat studies spot it.