Understanding Data – What you need to know

I like to follow this blog post because they cover the simple to the complex and everything in between. I found this article on there and I just have to share it because 1) it does a great job clarifying definitions, 2) it covers information that we all should know and 3) it’s an easy and enjoyable read.

You can read the entire post (and links to other great content) here: http://www.columnfivemedia.com/data-basics-need-know

The Data Basics You Need to Know

Data is everywhere these days, and it’s likely you do—or will—use it regularly. But we know it’s probably been a while since you sat through math class, or maybe you never quite learned the data basics. (Don’t worry, we won’t call you out.) Instead, we want to help you get a grasp so that you can better understand and ultimately visualize the data you work with. Here are the data basics you need to know.

WTF Is Data?

Data is any information you are collecting: numbers, statistics, measurements. It can also be words, observations, or other inputs.

DataTypeHierarchy

If you are dealing with numbers that represent something measurable, like sales of a product, you are dealing with quantitative data. If you are dealing with information that represents something less measurable, like how people feel about a product, you are dealing with qualitative data.

Each quantitative data point or variable you collect will be continuous or discrete, but as a whole, you are dissecting your data in one of two ways:

Cross-Sectional: The sample of elements is measured only once. This shows you a snapshot of variables at a point in time (e.g., market survey).

Customer Satisfaction

 A sample visualization of cross-sectional data. 

Longitudinal: The data sample is measured repeatedly over time (e.g., stock prices, monthly sales data).

Longitudinal data

 A sample visualization of longitudinal data collected over time. 

What Makes a Data Set?

A data set is comprised of variables; each individual data point—the thing that is measured or counted—is a variable. Each variable can be examined on its own or in relation to other variables to reveal insights, including:

Mean

Mean: The sum of all variables divided by the number of variables.

Range

Range: The difference between the highest and lowest variables in your data set.

Quantiles

Quantiles: The values taken at regular intervals from the inverse of the cumulative distribution function (CDF) of a random variable.

Deviation

Variability/Standard Deviation: Measures of how far a given variable is from the mean.

Deviation

Distribution: The distribution of data around a central value.

Outlier

Outliers: A variable that is an abnormal distance from other variables in your data set.

 

Data Relationships

Depending on what type of data you’ve collected, you will see different relationships represented in your data set. Understanding these relationships—and which visualizations communicate that relationship—will help you better communicate your data. Here are some of the most common. 

Nominal

Nominal comparison: This is a simple comparison of the quantitative values of subcategories (e.g., number of visitors to a website).

Chart Types for Nominal Comparison

Nominal-Comparison-Chart-Types

Timeseries

Time Series: This tracks change in value of a consistent metric over time (e.g., monthly sales).

Chart Types for Time Series

Time-Series Chart Types

Ranking

Ranking: This shows how two or more values compare to each other in relative magnitude (e.g., NBA players, ranked by height).

Chart Types for Ranking

Ranking-Chart-Type

PartToWhole

Part-to-Whole: This shows a subset of data compared to the larger whole. This is used to show things like proportion or percentages (e.g., percentage of customers purchasing various products).

Chart Types for Part-to-Whole

Part-to-Whole
Correlation

Correlation: This is data with two or more variables that may demonstrate a positive or negative correlation to each other (e.g., salaries by level of education).

Chart Type for Correlation

Scatterplot

Want to know more? Check out our lessons on pie charts, bar charts, and line charts.  

This post originally appeared on Visage.