You have data. You need insights. The bridge between "data" and actionable "insight" is often statistics.
But to a lot of us, stats are scary, foreboding monsters that remind us of a miserable 8am class sophomore year in college.
It's too bad, really, because learning stats in a vacuum - in a classroom without context - is like watching a sunset on TV. Nice, maybe, but not nearly as memorable or powerful as the real thing.
The real thing is in your data. It's hidden behind a set of deceptively simple concepts, tools and techniques.
So bring that data and let's explore The Zen of Data.
The Zen of Data is "stats for the dataphobic," also known as "stress free stats."
We operate on a few simple principles:
- You have data. And you have a spreadsheet. There's no need to grind out numbers.
- However, it's very important to understand the mathematical principles behind the way your data behaves. Or misbehaves, as the case may be.
- A gentle introduction to concepts + some practical tutorials in data operations (data functions in spreadsheets mainly, at this level) will take you a long way toward overcoming dataphobia.
That's what The Zen of Data is all about. You'll start with a gentle introduction to stats and data, then move right to practical, applied concepts around:
Relax! It's just data.
The Zen of Data provides a gentle introduction to statistics for data analytics. This video provides a gentle introduction to The Zen of Data and our approach to stats.
This course is designed for the "dataphobic" who is faced with the task of organizing, analyzing and especially acting on data.
This video - while admittedly a little more "laid back" than the regular course videos - gives an overview of "stress free stats."
Following along with the Intro video, this video provides a brief overview of the statistics concepts featured in the Zen of Data.
Stats provide the foundation of good analytics, and concepts like Average, Extrapolation, Variation and Analysis form the base of stats.
If you have data and need to dig in a little deeper, start here.
An average is just a number that represents other numbers. It provides a symbol for understanding the actual events and data behind it.
In this video, you will learn the 3 major approaches to "average," the ideas behind them and when to use them.
Don't worry. In The Zen of Data, everyone is above average!
Stats guru and co-instructor Mary Dereshiwsky shares her insights into averages and how they're used and calculated in terms that are easy to understand and apply.
We sift the Wild Wide Web to bring you premiere tutorials, walk-throughs and demonstrations of stats and analytic concepts.
They will enhance your understanding of the concepts we share and help you gain "fingers on" experience with applying them. Please enjoy, and let us know how we can help you create better analytics.
Calculating Mean, Median, Mode and Other Fancy Tricks in Excel
A great tutorial from a terrific YouTube instructor. Very lively, very useful step-by-step practical instruction to using Averages functions in Excel.
It's music you want, it's music you get!
A clever way to remember Mean, Median and Mode set to a tune that will be ringing in your head all day.
Variation exists in every part of our lives. Every system - human, biological, technical - experiences and exhibits variation.
Variation describes the tendency of things to be different. No matter what we measure, if we can look close enough, we will find variation.
Measure enough events and you'll get data. Study that data and you'll see variation up close and personal.
Variance is the statistical term, concept and calculation that describes and represents our effort to make mathematical sense of the way things vary.
In this lesson, we address the tendency for systems to vary and our approach to variance.
In our Variance lesson we focused on the idea that data tends to disperse from the center.
Understanding the nature of that spread - is it clustered close to the center or "distributed" way out into the digital hinterlands? - provides a key first step to doing good analytics.
Spreadsheet programs like Excel provide a simple yet profound way to calculate, analyze and "see" the nature of your distribution.
In this tutorial, the author reinforces our sense of data dispersal and provides some helpful tools for measuring variability.
If you look real close, it seems that data takes on a life of its own.
First, it tends to group around a center or "central tendency" (average).
Then, it simultaneously pulls away -varies - from the center (variance).
Now we're going to find that it disperses through a data set in an organized, predictable way.
It can all get kind of "quantum," if you're paying attention.
In this video, we begin exploring the nature of distribution - the tendency of data to disperse from the center - and emphasize probability as a way of describing the shape of your data.
Once we get a handle on the nature of variation in life and in data, it's natural to start looking at the way data disperses through a data collection.
In this video we finish exploring data's tendency to disperse like a peanut at an elephant convention.
Grouping data to show its distribution provides a sense of how the data centralizes and how it disperses.
This author uses a fairly simple data set to illustrate data grouping and distribution in Excel.
Our discussion of the Bell Curve provides a good structural understanding of distribution. However, deep in statistical science, the idea of the "normal distribution" holds sway.
"Normal" describes the even shape and scope of the Bell Curve - not skewed one way or the other - and the distribution of data in it.
We don't get too concerned about "normal" because - hey - why be normal!? Not really.
Actually, in our focus on practical stats for analytics we assume that you already have data. So whether or not it's "normal," it is what it is.
However, "normal" is still an important concept in statistical science and somebody who actually did stay awake in college stats might some day ask if the data in your distribution is "normal."
We always want you to be prepared! And a blank stare does not qualify as "prepared."
This tutorial provides a very useful discussion of "normal" in data distribution, since it's based on real data (stock returns) and explores the shape of reality over experimental design.
It's really a little bit spooky, isn't it, this tendency for data to centralize around an "average," yet disperse into "variance."
Well, beyond the cosmic absurdity of it all, it is indeed true that data not only centralizes and disperses, it does so in "standard" ways.
Standard deviation is one of those stats concepts that sounds vaguely menacing, but once you understand what it is and how it works, you start finding "standard deviations" all over your data.
Which, you would because it's, you know, a standard.
Although we present the concepts discretely - Averages, Variance, Distribution, Standard Deviation - because that makes them easier to ingest, in practical terms you need to use and understand all of them together.
This video tutorial does a terrific job of:
- Reviewing the key concepts
- Creating linkage between them (how does one impact or link to another?)
- Illustrating how to actually run the functions in Excel (2010, in this case)
We sift the Wild Wooly Web so you don't have to!
Note: There seems to be a problem with the embed on this video. I've added a couple other equivalent videos under Supplemental Materials. There should be a button/bar in the upper right hand corner of the panel. Apologies for the inconvenience. TW
Sometimes simple works best.
Step 1: Open Excel (this tutorial uses 2003 but, as it observes, the functions are pretty much the same across versions)
Step 2: Press "Play"
Step 3: Follow along step-by-step as the author guides you through STDEV (standard deviation function)
Genius-level stuff, really.
Hope you enjoy.
Most real world data challenges involve a sort of "dance" of variables. You have (at least) two factors arm wrestling for your attention. Both trying to convince you that they are the real reason for your good fortune!
Correlation is the stats concept behind the idea that sometimes things move, change or shift together. It gets us to a mathematical explanation for so many of the "if, then" conditions in our lives and in our data.
Todd Wieland operates Canis Learning Systems in Tampa, FL where he creates learning products and systems that help people leverage knowledge and technology to create great careers and businesses.
His projects focus on topics as diverse as marketing, analytics, manufacturing math, game and app development, filmmaking and media, automation and transportation technology.
The theme that ties all of Canis Learning Systems' and Todd's projects together is opportunity: We live in an age of unbounded possiblity, empowered and ennobled by the acquisition and application of knowledge.
The promise of the future for generations to come is built on specialization. Helping others gain and apply specialized knowledge, sharing specialized applications and leveraging unique niches is an economic and social trend with almost unlimited potential.
Coming from a long line of entrepreneurs and small business operators (he started working in his dad's neighborhood store at age 11), he has an eclectic background in education, technology, media and even runs a snow cone machine on the weekends.