Table of Contents
Graphs are everywhere and play a role in our lives whether we realize it or not
Not only do graphs allow us to make sense of complex data, but it is also nearly impossible for us to escape their influence. Every day we’re faced with new data points and graphs and it makes an impact on our lives. When people talk about how bad graphs are, it’s usually because they were poorly constructed or used a flawed metric. But there are some graphs that are just bad. This is one of them.
People have been making bad graphs since the beginning of time. The problem is that we tend to think that good graphs are pretty simple. We see a line, we assume it means something real, and then blindly apply it to our lives. The fact is that some graph problems are very difficult to solve, and sadly many of our favorite graphing libraries in use today fail us at this task.
The question is: how can we make sure that graphs remain useful in the future? The answer is to understand why they lie.
This article presents a series of real-world graphs, followed by explanations of what went wrong and suggestions on how to improve them. It should be stressed that some of these problems have been addressed before, but have fallen by the wayside with time. We believe that these findings are still relevant today. In fact, we believe this is one of the most important articles to ever be written about graphs.
Before plunging into the discussion, we must admit that there are a lot of graphs out there. This article is not intended to be anything more than a tour through the land of graphing. There are many other pages about graph theory and visualization, but we wanted to focus on illustrating how bad graphs can be.
Graphs can be used to track data, trends, and changes over time
As you can see, this graph uses dashes and dots to represent the difference between the number of people who smoke and don’t since 1995. The year 1996 is when the smoking ban started, so it makes sense to have that separation. What does not make sense is why since 1995 there have been three large gaps at around the same time. The graph tells a story of a sudden drop in smokers and then a gradual decline as if the timing of the ban had no effect.
Title:Graphs can be used to track data, trends, and changes over time
Graphs can be used to track data, trends, and changes over time [ARTICLE END]
Please use this article in your work. See the Copyright Policy for more information.
Product details from Barnes & Noble. This is what you’ll get:
This book is designed for college students. It’s a paperback edition, so it’s a little thicker than some of the other textbooks in this class. It also has thicker pages so it will be easier to write on than the thinner paper of some of the other textbooks. And since this book was published by a company that sells high quality textbooks and school supplies, this should not be an issue for you either.
They can help us make better decisions by providing visual representations of information
This graph shows the average weekly number of days lost due to winter weather from Jan 2011 to Dec 2012. In this case, you can see that there were only three weeks in total. It also shows that June and July 2011 were two of the warmest months on record, so it’s somewhat counterintuitive how much winter weather is going to hurt us during those two months. This is a better use of data than a table, but it’s still not the best way to communicate this information.
This second graph also uses only a table. This time, however, the data is arranged chronologically with the most recent years first and going backwards. The graph is broken up into four chunks of years, so it’s easier to read. The disadvantage is that you have to cross-reference between two tables and deal with the inherent inaccuracies of those kind of charts.
Sometimes graphs matter more than words when trying to explain something
Graphs can be easier to read and understand, especially when dealing with a lot of data. This is true even if the data points are not in a straight line or have outliers. The graph below shows that there’s a relationship between people watching television and their weight. There are certainly outliers, but the trend is there.
##
Title:Sometimes graphs matter more than words when trying to explain something
Sample “Quotes” : “The results suggested that people who watch television have a higher Body Mass Index (BMI)”
Graphs can be easier to read and understand, especially when dealing with a lot of data. This is true even if the data points are not in a straight line or have outliers. The graph below shows that there’s a relationship between people watching television and their weight. There are certainly outliers, but the trend is there.
We should take the time to understand graphs and how to read them correctly
The graph below is a good example of how complex a graph can be. It’s hard to visually interpret at first, but once you understand the relationships it might be easier to determine what is going on. The X-axis is a time interval and the Y-axis shows how often or how much people were eating between meals. The left three columns are for a group of people who were all eating every 2 hours and there are two groups on the right side, one group of people that was always eating every 2 hours, and another group that spent 2 hours very infrequently. The groups are called “Group 1” and “Group 2” in the table below.
As you can see, there are three different curves and each one seems to be less than the next. The top curve is for people who were eating two times per day, while the middle curve is for people that were eating three times per day. The other group of people was also eating three times per day, but they had a greater range where they would wait up to 2 hours before they had another meal.
Photo credit: http://www.dietdoctor.com/blog/why-your-body-is-craving-meals.html
