Descriptive Statistics cookbook

7 recipes · 43 steps · about 32 minutes of reading

Pick the average that represents a skewed column
My average keeps getting pulled off by a few big numbers, and it doesn't match what most of the rows look like.
BeginnerRead 4 minDoing: about 15 minutes6 steps
Your standard deviation doesn't match what your teammate's tool reports
I calculated the standard deviation two different ways and got two different numbers, and I don't know which one is right.
BeginnerRead 3 minDoing: twenty minutes6 steps
An outlier is dragging your summary numbers off track
One weird row is throwing off my average and I can't tell if it's a mistake or a real result.
Beginner to intermediateRead 6 minDoing: 20 to 30 minutes6 steps
Rule out a hidden cause before calling a correlation a driver
Two numbers move together on my chart and my manager wants to know if one is causing the other.
Beginner to intermediateRead 5 minDoing: 30 to 45 minutes for a first pass, quicker once the query is saved6 steps
Pull a fair sample from a huge table for a quick check
My table has millions of rows, and I need a quick, honest look at it without pulling the whole thing.
Beginner to intermediateRead 5 minDoing: 20 to 30 minutes6 steps
Compute percentiles in SQL for a table too big for a spreadsheet
I need the median and the 90th percentile of a table with ten million rows, and Excel can't even open the file.
IntermediateRead 4 minDoing: 20 to 30 minutes for a first pass, quicker once the query is saved6 steps
Measure a column's skew to know if the average can be trusted
I'm about to hand off a summary of this column and I have no idea if it's symmetric, lopsided, or has two humps.
IntermediateRead 5 minDoing: about 20 minutes7 steps

Cookbook badge

Finish all 7 recipes to earn this.

Descriptive Statistics cookbook

Pick the average that represents a skewed column

Your standard deviation doesn't match what your teammate's tool reports

An outlier is dragging your summary numbers off track

Rule out a hidden cause before calling a correlation a driver

Pull a fair sample from a huge table for a quick check

Compute percentiles in SQL for a table too big for a spreadsheet

Measure a column's skew to know if the average can be trusted

In these roadmaps

Descriptive Statistics cookbook

Pick the average that represents a skewed column

Your standard deviation doesn't match what your teammate's tool reports

An outlier is dragging your summary numbers off track

Rule out a hidden cause before calling a correlation a driver

Pull a fair sample from a huge table for a quick check

Compute percentiles in SQL for a table too big for a spreadsheet

Measure a column's skew to know if the average can be trusted

In these roadmaps