Data Science Fact

Daily data science tweets from @JohnDCook.

[Data Science Fact @DataSciFact]

“Statistics is hard, especially when effects are small and variable and measurements are noisy. There are no quick fixes … and a formulaic approach to statistics is a principal cause of the current replication crisis.” — Blakeley McShane

[Data Science Fact @DataSciFact]

Modes of convergence diagram https://t.co/QQeqMEC3r2

[Data Science Fact @DataSciFact]

The word 'statistics' comes from 'state' because first applications were for governments.

[Data Science Fact @DataSciFact]

Unjustified assumptions of independence are the root of statistical evil.

[Data Science Fact @DataSciFact]

'BLUE' stands for best linear unbiased estimator, 'best' in the sense of mean squared error.

[Data Science Fact @DataSciFact]

Visualizing correlations with graphs https://t.co/RdRT9fbjtx

[Data Science Fact @DataSciFact]

'The language of science is the language of probability, and not of p-values.' -- Luis Pericchi

[Data Science Fact @DataSciFact]

An ordinary least-squares regression line passes through the point (mean x, mean y).

[Data Science Fact @DataSciFact]

A quantile-quantile plot is commonly used as a quick way to check to tell whether data appear to be normally distributed.

[Data Science Fact @DataSciFact]

Balanced incomplete block designs (BIBDs)

https://t.co/l8ViK9NUeA

[Data Science Fact @DataSciFact]

'The knowledge of an average is a meager piece of information.' -- Francis Galton

[Data Science Fact @DataSciFact]

V-statistics https://t.co/As4XoD1hEg

[Data Science Fact @DataSciFact]

Education studies are hardly ever replicated https://t.co/NVYUXh2wes

[Data Science Fact @DataSciFact]

Bayes factors were introduced independently by Alan Turing and Harold Jeffreys.

[Data Science Fact @DataSciFact]

Subject to mild regularity assumptions, maximum likelihood estimators are asymptotically efficient.

[Data Science Fact @DataSciFact]

Dump a pickle file to a readable text file https://t.co/qoyARu6h4E

[Data Science Fact @DataSciFact]

Drawbacks of k-means https://t.co/sOqaPjy0Pe

[Data Science Fact @DataSciFact]

Estimating the Gini coefficient of a population form the Gini coefficients of subpopulations https://t.co/qrLwSUiR9y

[Data Science Fact @DataSciFact]

'Data isn’t fissile material. It doesn’t spontaneously reach critical mass and start producing insights.' -- Marko Karppinen

[Data Science Fact @DataSciFact]

Utility functions are contentious so we agree on a mediocre implicit utility rather than pick a better explicit one.

[Data Science Fact @DataSciFact]

Can you have confidence in a confidence interval? https://t.co/6WahD6K2Cc