Sample Size & Power Analysis: How Many Observations Do You Need?
Plan studies, surveys and experiments in Excel: margin of error and sample size for means and proportions, the finite population correction, Cohen's d and the MDE, two-group sample sizes with NORM.S.INV, power curves, power and MDE for a fixed n, and Bonferroni, unequal allocation, dropout and non-response, in 8 graded labs.
The first lesson is free — no account needed.
About this course
A sample size and power analysis course that plans the studies of one fictional research firm, Larkspur Analytics, in a spreadsheet, for analysts, researchers, students writing a thesis, product teams and anyone asked how many observations they need. You find the margin of error and the sample size for a mean, size a survey for a proportion and shrink it with the finite population correction, turn a pilot into a pooled standard deviation, Cohen's d and a minimum detectable effect, calculate n per group for comparing two means with NORM.S.INV and a small-sample correction, size an A/B test for two proportions, draw power curves in a scatter chart, find the power and MDE of a fixed budget, and adjust for multiple comparisons, unequal groups, dropout and non-response. Every formula is built by hand, so you see what a sample size calculator does. The first lab is free.
Lessons
- Margin of Error & Sample Size for a MeanFree
- Sample Size for a Proportion & the Finite Population Correction Pro
- Effect Size: Pooled SD, Cohen's d & the Minimum Detectable Effect Pro
- Sample Size for Comparing Two Means (NORM.S.INV) Pro
- Sample Size for Two Proportions: Sizing an A/B Test Pro
- The Power Curve: Power vs Sample Size (Scatter Chart) Pro
- Power and Minimum Detectable Effect for a Fixed Sample Size Pro
- Practical Adjustments: Bonferroni, Unequal Allocation, Dropout & Non-Response Pro