Data sources
Every dataset behind the site and the app, with its license and citation.
Every chart on this site carries a label saying what it is: a live calculation on real data, a recorded run, or a simulation. This page says where the data came from.
Charts on this site
| Figure | Data | License | Label |
|---|---|---|---|
| Hero, Analyst | Palmer penguins, raw flipper length and body measurements | CC0-1.0 | Live on real data |
| One result | ToothGrowth | R datasets package, published results | Recorded result |
| Workflow: profile, check, compare, model, explain, forecast | New York air quality | R datasets package, public records | Recorded result, and for the model comparison "Recorded runs, test settings, not tuned" |
| Statistician | Simulated interval-censored survival data, written for CodeSplash Analytics | Our own simulation | Recorded on simulated data |
| Data scientist, ML engineer | Palmer penguins | CC0-1.0 | Recorded result |
| Parity strip | tests/parity/report.json, our own test report | Our own report | Test report |
| NIST | NIST StRD certified values, tests/nist/report.json | Public domain | Test report |
| Academy power explorable | A simulation that runs in your browser | Our own model | Live simulation |
Bundled with the app
These four datasets come with the app for lessons and examples.
New York air quality, 1973
Daily air-quality readings in New York from May to September 1973: ozone, solar radiation, wind and temperature. Real missing values (ozone is missing on 37 days) make it a good first dataset for missing data, skewed measures and simple regression.
- License
- R-datasets. Shipped with R's datasets package as airquality; the readings come from the New York State Department of Conservation and the US National Weather Service (public records).
- Citation
- Chambers JM, Cleveland WS, Kleiner B, Tukey PA (1983). Graphical Methods for Data Analysis. Belmont, CA: Wadsworth.
Messy customer list
A simulated customer export with every common data problem: sentinel values like -99, mixed date formats, stray whitespace and case in categories, inconsistent yes/no codes, duplicates and missing values. Built to trip each of the Prepare page's suggested fixes.
- License
- Simulated. Generated for CodeSplash Analytics; no real people.
- Citation
- CodeSplash Analytics test data (tests/recipes/fixtures/messy_customers.csv).
Palmer penguins
Measurements of 344 penguins of three species (Adélie, Chinstrap, Gentoo) on three islands of the Palmer Archipelago, Antarctica, 2007–2009. The Academy's recurring dataset: groups to compare, measurements that correlate, missing values to handle and species to classify.
- License
- CC0-1.0
- Citation
- Horst AM, Hill AP, Gorman KB (2020). palmerpenguins: Palmer Archipelago (Antarctica) penguin data. Data collected by Dr. Kristen Gorman and the Palmer Station LTER. Source
Tooth growth in guinea pigs
The length of odontoblasts (tooth-growing cells) in 60 guinea pigs, each given vitamin C at one of three doses (0.5, 1 or 2 mg/day) by one of two delivery methods: orange juice (OJ) or ascorbic acid (VC). A small, clean experiment for comparing groups.
- License
- R-datasets. Shipped with R's datasets package as ToothGrowth; the measurements are published experimental results (Crampton 1947), and facts carry no copyright.
- Citation
- Crampton EW (1947). The growth of the odontoblast of the incisor teeth as a criterion of vitamin C intake of the guinea pig. The Journal of Nutrition 33(5): 491–504.
Worked examples
The Academy and the worked examples use small published datasets. Each is credited to its source below. The rest of the worked-example data is simulated or was written for CodeSplash Analytics. The app's Settings, under About, has the same list in its third-party notices.
All 117 datasets
| Dataset | Credit |
|---|---|
a_depression | Hollander & Wolfe (1973) depression scores, first and second visit (stats::wilcox.test example) |
airquality | datasets::airquality (has missing values) |
aml | survival::aml (Miller 1997) |
approval | Agresti, A. (2013). Categorical Data Analysis (3rd ed.), Table 11.1 |
attitude | datasets::attitude (Chatterjee & Price 1977) |
biopsy | MASS::biopsy (Wolberg & Mangasarian 1990) |
birthwt | MASS::birthwt (Hosmer & Lemeshow 1989); race recoded to labels, 0/1 columns as booleans |
c1_mpdta | did::mpdta — county teen employment 2003–2007, minimum-wage increases (Callaway & Sant'Anna 2021) |
c1_senate | rdrobust::rdrobust_RDsenate (Cattaneo, Frandsen & Titiunik 2015) |
c2_lalonde | MatchIt::lalonde (LaLonde 1986; Dehejia & Wahba 1999) |
c3_store_brand | Conover (1999) store x brand sales (stats::quade.test example), long format |
cb_anscombe | datasets::anscombe (Anscombe, F. J. (1973). Graphs in statistical analysis. The American Statistician, 27, 17-21) |
cb_challenger | vcd::SpaceShuttle (Dalal, S. R., Fowlkes, E. B., & Hoadley, B. (1989). Risk analysis of the Space Shuttle: pre-Challenger prediction of failure. JASA, 84, 945-957, Table 1; Presidential Commission report, 1986) |
cb_datasaurus | datasauRus::datasaurus_dozen_wide (MIT; Matejka, J., & Fitzmaurice, G. (2017). Same stats, different graphs. CHI 2017; dinosaur by Alberto Cairo) |
cb_galton | HistData::Galton (Galton, F. (1886). Regression towards mediocrity in hereditary stature. Journal of the Anthropological Institute, 15, 246-263) |
cb_kidney_stones | Charig, C. R., Webb, D. R., Payne, S. R., & Wickham, J. E. A. (1986). Comparison of treatment of renal calculi by open surgery, percutaneous nephrolithotomy, and extracorporeal shockwave lithotripsy. BMJ, 292, 879-882, Table I (one row per patient) |
chickwts | datasets::chickwts (six groups) |
co2_long | datasets::CO2 (Potvin, Lechowicz & Tardif 1990) |
f7_dax | datasets::EuStockMarkets, DAX, last 750 trading days |
f7_seatbelts | datasets::Seatbelts (Harvey & Durbin 1986); petrol price × 100 |
f7_seatbelts_future | datasets::Seatbelts plus twelve future rows with known regressors |
h_faithful | datasets::faithful |
hair_eye | datasets::HairEyeColor expanded to one row per student |
housing | MASS::housing (Madsen 1976), one row per resident (Freq expanded) |
insurance | MASS::Insurance (Baxter, Coutts & Ross 1980) |
iris | datasets::iris |
lb1_crins | bayesmeta::CrinsEtAl2014 (Crins et al. 2014, Pediatric Transplantation 18, 839–850) |
life_aircon | Proschan (1963) air-conditioning failure intervals, as boot::aircondit and boot::aircondit7. |
life_bearings | Lieblein & Zelen (1956), ball-bearing fatigue life (millions of revolutions). |
liver_scan | Altman, D. G., & Bland, J. M. (1994). Diagnostic tests 1: sensitivity and specificity. BMJ, 308, 1552. |
lung | survival::lung (Loprinzi et al. 1994) |
m1_assink | metadat::dat.assink2016 (Assink & Wibbelink 2016) |
m1_bcg | metadat::dat.bcg (Colditz et al. 1994) |
m1_hackshaw | metadat::dat.hackshaw1998 (Hackshaw et al. 1997) |
m1_molloy | metadat::dat.molloy2014 (Molloy et al. 2014) |
m1_multilevel | metadat::dat.konstantopoulos2011 (Konstantopoulos 2011) |
m1_normand | metadat::dat.normand1999 (Normand 1999) |
m1_raudenbush | metadat::dat.raudenbush1985 (Raudenbush 1984) |
malformation | Graubard, B. I., & Korn, E. L. (1987). Biometrics, 43, 471–476; Agresti (2013) Table 3.7 |
mendel_peas | Mendel, G. (1866). Versuche über Pflanzen-Hybriden. |
mgus2_cr | survival::mgus2 (Kyle et al. 2002), competing-risks coding as in the survival vignette |
ml_airpassengers | datasets::AirPassengers (Box & Jenkins 1976) |
ml_airquality | datasets::airquality (New York, May–September 1973) |
ml_chickweight | datasets::ChickWeight (Crowder & Hand 1990) |
ml_deaths | datasets::mdeaths, datasets::fdeaths (Diggle 1990) |
ml_digits | UCI Optical Recognition of Handwritten Digits, test set (Alpaydin & Kaynak 1998; CC BY 4.0) via sklearn.datasets.load_digits: first 100 images per digit |
ml_iris | datasets::iris (Anderson 1935; Fisher 1936) |
ml_nile | datasets::Nile (Durbin & Koopman 2001) |
ml_penguins | datasets::penguins (R 4.5; Horst, Hill & Gorman 2020) |
ml_pima | MASS::Pima.tr + MASS::Pima.te (Smith et al. 1988) |
ml_pima_rare | MASS::Pima.tr + MASS::Pima.te, all negatives and the first 40 positives (Smith et al. 1988) |
ml_titanic | datasets::Titanic expanded to one row per passenger (Dawson 1995) |
ml_titanic_tx | datasets::Titanic as one transaction per passenger |
ml_ukgas | datasets::UKgas (Durbin & Koopman 2001) |
ml_usarrests | datasets::USArrests (McNeil 1977) |
mtcars | datasets::mtcars (plus a column name with spaces and a boolean) |
mv_baumann | carData::Baumann (Baumann & Jones; Fox & Weisberg 2019) |
mv_dune | Dutch dune meadows: cover of 30 plant species at 20 sites with management (Jongman, ter Braak & van Tongeren 1987; vegan::dune) |
mv_mite | Oribatid mites in 70 moss cores with coordinates (m) and environment (Borcard & Legendre 1994; vegan::mite, mite.xy, mite.env) |
mv_obrienkaiser | carData::OBrienKaiser (O'Brien & Kaiser 1985) |
mv_pottery | carData::Pottery (Tubb, Parker & Nickless 1980) |
mv_vare | Lichen pastures in Finland: cover of 44 species and 14 soil variables at 24 sites (Väre, Ohtonen & Oksanen 1995; vegan::varespec, varechem) |
panel_empluk | Arellano & Bond (1991); R package plm (dataset EmplUK). Unbalanced. |
panel_grunfeld | Grunfeld (1958); R package plm (dataset Grunfeld). |
panel_produc | Munnell (1990); R package plm (dataset Produc). |
pefr | Bland, J. M., & Altman, D. G. (1986). The Lancet, 327, 307–310 (first readings) |
pima | MASS::Pima.tr and MASS::Pima.te (Smith et al. 1988) |
plantgrowth | datasets::PlantGrowth |
q1_circuit | qcc::circuit (Montgomery 2013, Table 7.7); Units added for a u chart |
q1_filtration | Montgomery (2017), Design and Analysis of Experiments, Example 6.2 |
q1_nile | datasets::Nile (Durbin & Koopman 2001) |
q1_orangejuice | qcc::orangejuice (Montgomery 2013, Table 7.1) |
q1_pistonrings | qcc::pistonrings (Montgomery 2013, Introduction to Statistical Quality Control, Table 6.3) |
quine | MASS::quine (Quine 1975) |
rounding_times | Hollander & Wolfe (1973) rounding-first-base times (stats::friedman.test example), long format |
s1_cbpp | lme4::cbpp (Lesnoff et al. 2004) |
s1_dietox | geepack::dietox (Lauridsen et al. 1999), sorted by pig and week; 2 weights set missing |
s1_mcycle | MASS::mcycle (Silverman 1985) |
s1_ohio | geepack::ohio (Fitzmaurice & Laird 1993), sorted by child and age |
s1_pastes | lme4::Pastes (Davies & Goldsmith 1972): casks nested within batches |
s1_penicillin | lme4::Penicillin (Davies & Goldsmith 1972): plates crossed with samples |
s1_sleep | lme4::sleepstudy (Belenky et al. 2003); sleep group = alternate subjects; 3 reaction times set missing |
s1_survey | MASS::survey (Venables & Ripley 2002), renamed columns |
s1_ticks | lme4::grouseticks (Elston et al. 2001); HEIGHT standardized |
s2_bladder | survival::bladder2 (Byar 1980; Wei, Lin & Weissfeld 1989) |
s2_cgd | survival::cgd (Fleming & Harrington 1991) |
s2_heart | survival::heart (Crowley & Hu 1977) |
s3_hs | lavaan::HolzingerSwineford1939 (Holzinger & Swineford 1939) |
s3_kripp | Krippendorff (2011), Computing Krippendorff's alpha-reliability (= irrCAC::cac.raw4raters) |
s3_lsat7 | mirt::LSAT7 expanded (Bock & Lieberman 1970) |
s3_poldem | lavaan::PoliticalDemocracy (Bollen 1989) |
s3_science | mirt::Science (Karlheinz & Melich 1992) |
s4_bio | pscl::bioChemists (Long 1990): articles in the last 3 years of the PhD, 915 students |
s4_coffee | Agresti (2002) Table 10.3: coffee brand choice at two purchases (one empty off-diagonal cell) |
s4_drugs | Agresti (2002) Categorical Data Analysis, Table 12.1 (favourable response to three drugs) |
s4_haireye_counts | datasets::HairEyeColor as one row per cell with its count (32 rows) |
s4_job | fisher.test example: job satisfaction by income (Agresti 2002, Table 2.8), 96 workers |
s4_nhanes | mice::nhanes2 (Schafer 1997): 25 people, missing BMI, hypertension and cholesterol |
s4_swiss | datasets::swiss (1888 Swiss provinces), 47 rows |
s4_vision | Stuart (1955) Biometrika 42: unaided distance vision, 7477 women |
s5_risk | MASS::Pima.tr + MASS::Pima.te (Smith et al. 1988); predictions of two logistic models fitted on them |
shrout_fleiss | Shrout, P. E., & Fleiss, J. L. (1979). Psychological Bulletin, 86, 420–428, Table 2 |
sleep_long | datasets::sleep (Student 1908), long format |
sleep_wide | datasets::sleep (Student 1908), reshaped wide by ID |
sp_cells | Crick & Ripley's biological cells, spatstat.data::cells; a regular pattern on the unit square. |
sp_columbus | Columbus OH neighbourhoods 1980 (Anselin 1988), spData::columbus (CC0). |
sp_japanesepines | Numata's Japanese black pines, spatstat.data::japanesepines; unit square. |
sp_meuse | Meuse river topsoil zinc (Rikken & Van Rijn 1993), sp::meuse; planar metres. |
sp_nc_sids | North Carolina SIDS 1979 by county (Cressie 1993), from the sf package's nc.shp. |
sp_redwood | Strauss's redwood seedlings, spatstat.data::redwood; a clustered pattern on [0, 1] × [−1, 0]. |
survey | MASS::survey (Venables & Ripley 2002), selected columns; exercise and smoking in natural order |
toothgrowth | datasets::ToothGrowth |
ts_co2 | datasets::co2 (Mauna Loa, Keeling) |
ts_lakehuron | datasets::LakeHuron (Brockwell & Davis 1991) |
ucb_admissions | datasets::UCBAdmissions expanded to one row per applicant |
veteran | survival::veteran (Kalbfleisch & Prentice 1980) |
warpbreaks | datasets::warpbreaks (ties; ordered tension L < M < H) |