Data Modelling
for Science
Companion tools and quizzes for the module, listed by section and lecture.
Classify That Variable
Sort each variable into Nominal, Ordinal, Discrete or Continuous. Hints on request, an explanation after every answer, and a breakdown by type at the end.
Start →Histogram Bin-Width Explorer
Drag a slider to rebin the CK-concentration data and watch the shape appear, flatten and fragment, while the skewness never budges.
Open →Random Variables
How a rule turns an outcome space into a distribution, worked through twice, with two dice and with a Punnett square.
Open →Independent and Mutually Exclusive Events
Decide whether each pair of events is independent, mutually exclusive, or neither, with the full deck on screen so you can check the counts.
Start →Adding Up Probabilities
Select the values a question asks about and watch their probabilities add up, with preset questions from the lecture.
Open →Die-Roll Convergence
Roll a die a few times, then a few hundred, and watch the relative frequency of any face settle towards 1/6.
Open →Snapdragon and Class Height Example
Two experiments set out in the same framework, a snapdragon cross and a class of 150 heights, each followed by worked probability questions.
Open →Is It Binomial?
Six scenarios. For each one, decide which of the four Binomial conditions hold before seeing whether the setting is Binomial.
Start →Build the Binomial Formula
Derive the Binomial formula step by step, from listing every sequence of successes and failures to the final probability.
Open →Binomial Experiment Lab
Run a Binomial experiment thousands of times and watch the observed frequencies settle onto the theoretical distribution.
Open →PMF vs PDF Explorer
Probability as the height of a bar for a discrete variable, and as the area under a curve for a continuous one.
Open →The Sampling Table
Four ways to choose k items from n, depending on whether order matters and whether repeats are allowed, with every outcome listed.
Open →Binomial Table Reader
Rewrite any Binomial probability into the cumulative form the tables give, then read the answer off the table step by step.
Open →Normal Shape Explorer
Move μ and σ to see how the mean sets the location of a Normal curve, and how the standard deviation sets both its width and its height.
Open →The Empirical Rule
Shade the area within one, two or three standard deviations of the mean of a Normal distribution, and compare the exact areas with the 68, 95 and 99.7 rule.
Open →Normal Table Reader
Standardise any Normal probability, rewrite it into the form Table 4 gives and read it off, or work backwards from a percentage with Table 5.
Open →Sampling Distribution of a Proportion
Draw repeated random samples and watch the sample proportion vary around the fixed population proportion, building its sampling distribution.
Open →Normal Approximation to the Binomial
Lay a Normal curve over Binomial bars, check when n is large enough, and compare the approximate probability, with or without a continuity correction, against the exact one.
Open →The Central Limit Theorem
Draw repeated samples from populations of different shapes and watch the distribution of the sample mean approach the Normal curve that the Central Limit Theorem predicts.
Open →Confidence Interval Simulator
Draw repeated samples, build a confidence interval from each one, and count how often the intervals contain the fixed population mean.
Open →NHANES Stats Workbench
Every idea from Lectures 01 to 12, from types of data to confidence intervals, worked on one real health survey of 7,235 US adults, where the true population values are known.
Open →Inference Visualisation
Follow the in-class R script from waiting times at a bus stop to the Central Limit Theorem, then on to standardising, z versus t, testing and confidence interval coverage.
Open →Hypothesis Testing
The idea and the five steps of a test for one mean, worked on a sugar claim with a live t curve, a simulation of the p-value and a short quiz.
Open →