Post-class Insights · Demo data
Predicted vs. observed difficulties
Lecture 3: Measuring Spread · Introduction to Statistics · Fictional demonstration
Key learning points
- The mean alone hides how spread out data is.
- Range (max − min) is quick but sensitive to outliers.
- Variance averages the squared deviations from the mean; for a sample, divide by n − 1.
- Standard deviation is the square root of variance, in the original units.
- For roughly normal data, about 68% of values lie within 1 SD of the mean.
Important formulas
s² = Σ(xᵢ − x̄)² / (n − 1)
s² sample variance, xᵢ each value, x̄ sample mean, n number of values
s = √s²
s sample standard deviation
Hardest concepts
- Why we divide by n − 1 (Bessel's correction, degrees of freedom).
- What 'unbiased estimator' means.
Preparation for the next lecture
Problem sheet Q1–4 before Thursday's tutorial. Next lecture covers estimators, which explains n − 1 properly.
Prediction comparison · Simulated data
Supplied demo summary: “ClassTwin predicted 3 of the 4 real confusion hotspots before the lecture.”
“Real” in the supplied summary refers to fictional observations in this demo, not actual student feedback or live ClassTwin results.
Predicted-and-observed matches
Ballpark figure (00:32)
Predicted: high (Mei) · 2 observed demo questions
Why n − 1 (01:08)
Predicted: very high (4 of 5 students) · 8 observed demo questions
68% rule (02:20)
Predicted: medium (Mei, Aisha) · 2 observed demo questions
Unexpected difficulties
Squared units (01:40)
Predicted: low · 2 observed demo questions
Teaching action
Tom's persona will be updated: squaring the units also confuses weak-maths students.
Predicted but not reported
No comparison cases in this category were supplied.
A predicted difficulty with no recorded question does not necessarily mean students understood it. No such comparison row was supplied; no “easier than predicted” conclusion is inferred.