10 km race results

Results of 500 finishers in a fictional community 10 km road race organised by the Corniche Striders running club, with each runner's self-reported weekly training distance. Each row is one finisher: age, age group, sex, training km per week and finish time. Built for correlation and regression in a sports context — training volume is strongly (negatively) related to finish time, but the data are observational.

Open in the lab Download CSV

Rows: 500
Columns: 6
One row is: one race finisher
Version: v1

Data dictionary

Column	Type	Unit	Summary	Description
`bib` Bib number	id			Race bib number.
`age` Age	integer	years	mean 38.81, median 37, SD 11.65, range 20–78	Age on race day.
`age_group` Age group	ordinal		5 levels: 18-29, 30-39, 40-49, 50-59, 60+	Age category used for prizes.
`sex` Sex	category		2 levels: Female, Male	Sex category of the entry.
`training_km_week` Weekly training	number	km/week	mean 23.71, median 20.85, SD 14.99, range 1.7–102.3	Self-reported average running distance per week over the previous 8 weeks.
`finish_min` Finish time	number	minutes	mean 57.8, median 57.6, SD 8.359, range 35.3–86.3	Chip time for 10 km.

What it is good for

Describe the relationship between training volume and finish time with a scatter plot and correlation.
Notice the curved (diminishing) relationship and compare Pearson and Spearman correlations.
Fit a regression of finish time on training, age and sex and interpret it.
Explain why the association need not be causal (fitter people may train more).

Methods: scatter plot, Pearson correlation, Spearman correlation, simple regression, multiple regression, log transformation.

Data quality notes

Training volume is self-reported and right-skewed.
The training–time relationship is curved: extra kilometres help less for runners who already train a lot.
No missing values.

Provenance and licence

Age ~ 18 + gamma (median ≈37). Training km/week is gamma-distributed (median ≈20) and slightly higher for runners in their 30s–40s. Finish time (minutes) = (68 − 13·ln(1 + training/10) + 0.35 per year of age beyond 35 + 6 for women) × log-normal individual variation (sd 0.08), bounded below at 31 minutes.

Synthetic data · CC0-1.0 (synthetic data generated by Statistics Quest) · generated deterministically (seed 20241019). It does not describe real people or organisations; any resemblance is coincidental.

Preview

Dataset

10 km race results

500 rows · one row = one race finisher · synthetic data

About this dataset Open in the lab

Bib number	Age (years)	Age group	Sex	Weekly training (km/week)	Finish time (minutes)
B001	30	30-39	Female	17	65.3
B002	41	40-49	Male	28.2	52.5
B003	28	18-29	Male	30.1	56.3
B004	24	18-29	Male	6.1	70.2
B005	22	18-29	Male	12.6	61.8
B006	24	18-29	Male	54.1	45.8
B007	36	30-39	Male	32.1	48
B008	43	40-49	Male	12.3	58.3
B009	25	18-29	Male	27.6	59.6
B010	57	50-59	Male	16.5	66.2
B011	37	30-39	Female	9.3	61.5
B012	33	30-39	Female	35.4	64.5

Used in

z-scores and relative standing · Describing Data
The normal distribution · Probability Distributions
Standardising and normal calculations · Probability Distributions
The t, chi-square and F distributions · Probability Distributions
Quantiles and inverse probability · Probability Distributions
Choosing and checking a distribution · Probability Distributions
Covariance and correlation · Correlation and Regression
Rank correlation · Correlation and Regression
Residual analysis · Correlation and Regression
Outliers, leverage and influence · Correlation and Regression
Does training volume pay off? · Case