A lot of us create graphs the way we choose dish soap at the grocery store, without much thought and based on habit. In reality, the final product (a beautiful and insightful graph) deserves more thought than that. First, it’s good to think about the end goal in communicating data. Do you want to show a trend over time, show discoveries by category, or maybe visualize variables in your data as a part of a whole? DataHero is great at suggesting the right chart for your data, but in this post you’ll find some background on why.
These two different graphs can seem nearly interchangeable but generally, line graphs work best for continuous data, whereas bar and column graphs work best for categorical data. Remember, there are exceptions to every rule, but these are some broad rules of thumb. Continuous data is quantitative, you cannot count the number of different values. This includes data like sales, height, profit, etc. It can also include time, though time can be both continuous and categorical data. Bar and column graphs are great representations of categorical data, in which you can count the number of different categories. For this example, time is continuous in the line graph. However, it can also be categorical in the bar graphs with categories of wine.
You’ll see below that the line graph represents the trend fairly well, but may be best for the total sale of wines (table, dessert and sparkling).
Here is a grouped bar chart of this same data broken down by category (table, dessert and sparkling). With grouped charts it can be difficult to tell the difference between totals in each group however it’s ideal to compare between each element in categories.
Stacked bar charts are great for showing a total, however it can be hard to compare the sizes of individual categories.
A percentage bar chart represents how each category makes up a part of the whole. In other words, it hides the quantity and represents the relative difference between quantities in each group. If there is an overlap in categories, it’s necessary to use a grouped bar chart, as there could be no part to whole relationship.
Cumulative bar charts represent the quantity of a variable as a sum of all frequencies up to that point. This would be ideal if I need to know how much dessert wine has been sold from 1994 to 2012, for example.
A logarithmic bar chart responds to skewed data, especially if one or a few points are skewed larger than the bulk of the data. It shows percentage change or multiplicative factors.
Find more wine statistics here. All these graphs use the same exact data but show vastly different perspectives, depending on what needs to be communicated from the data. DataHero allows you to easily navigate between all these graphs with a simple click of the “More” button, no extra functions or formulas required.
DataHero helps you unmask the answers in your data. There’s nothing to download or install. Simply create an account and connect to the data services you use every day (like Salesforce, Stripe, MailChimp, and Google Drive). DataHero automatically decodes your data and shows you the answers you need through dynamic visualizations.
Get the fastest, easiest way to understand your data today.Sign up for free