Data Visualization

Visualizing Your Data

Numbers in a table can tell you a lot, but a good graph can tell a story. Data visualization turns spreadsheets into insights, revealing patterns, outliers, and trends that are easy to miss in raw data. In Stata, creating graphs is a straightforward process that builds directly on the command syntax you've already learned. The main command is graph, and from there, you can specify the type of chart you want to create.

Creating Basic Graphs

Let's start with a few of the most common graph types. We'll use Stata's built-in auto.dta dataset for these examples. You can load it by typing sysuse auto.

A histogram is perfect for understanding the distribution of a single continuous variable. It shows how frequently different values occur in your dataset.

/* 
  Generate a histogram for the 'price' variable. 
  This shows us the distribution of car prices.
*/
histogram price

A scatter plot is the best way to see the relationship between two continuous variables. Each point on the graph represents one observation, plotted according to its value on the x and y axes.

/*
  Create a scatter plot to see the relationship
  between miles per gallon (mpg) and price.
  The first variable listed goes on the y-axis.
*/
graph twoway scatter price mpg

When you want to compare a summary statistic (like the average) across different categories, a bar chart is your best tool.

/*
  Create a bar chart showing the mean price
  for domestic and foreign cars.
  'over(foreign)' splits the data by category.
*/
graph bar (mean) price, over(foreign)

Customizing Your Visuals

Stata's default graphs are functional, but you'll often want to tweak them to make them clearer and more professional. You can do this by adding options to your graph commands, just like with other Stata functions. These options are added after a comma.

Participants will learn how to create different types of plots (scatter plots, bar charts, histograms, boxplots).

Let's improve our scatter plot by adding titles for the graph and its axes. Clear labels are essential for anyone trying to understand your chart.

/*
  Add a main title and labels for the x and y axes.
*/
graph twoway scatter price mpg, \
  title("Car Price vs. Fuel Efficiency") \
  ytitle("Price (in USD)") \
  xtitle("Miles Per Gallon")

You can also change the appearance of the graph's elements, like the color and shape of the markers on a scatter plot.

/*
  Change marker color to blue and shape to triangles.
*/
graph twoway scatter price mpg, \
  title("Car Price vs. Fuel Efficiency") \
  ytitle("Price (in USD)") \
  xtitle("Miles Per Gallon") \
  mcolor(blue) msymbol(triangle)

Interpreting Graphical Outputs

Creating a graph is only half the job. The real skill is in interpreting what it shows. Each type of graph highlights different aspects of your data.

When looking at a scatter plot, ask yourself: Is there a relationship? As one variable increases, does the other tend to increase (positive correlation) or decrease (negative correlation)? Or is there no clear pattern? Also, look for outliers, which are points that fall far from the general trend.

With a histogram, examine the shape. Is the data symmetric, like a bell curve? Is it skewed, with a long tail to one side? Where is the center of the data clustered?

For a bar chart, the interpretation is usually more direct. Compare the heights of the bars. Which category has the highest average? The lowest? How large is the difference between them?

Best Practices for Visualization

A great visualization is clear, honest, and easy to understand. Here are a few guidelines to follow.

Guideline	Why It Matters
Label Everything Clearly	Your audience shouldn't have to guess what the axes, title, or data points mean.
Choose the Right Graph	A bar chart can't show a distribution, and a histogram can't compare category means. Use the right tool for the job.
Keep it Simple	Avoid clutter. Too many colors, 3D effects, or unnecessary labels can distract from the main message.
Use a Consistent Scale	Always start your y-axis at zero for bar charts to avoid misleading comparisons. Keep scales consistent when comparing multiple graphs.

Following these principles helps ensure that your graphs accurately represent your data and effectively communicate your findings.

Quiz Questions 1/5

What is the primary command in Stata for creating all types of graphs?

Quiz Questions 2/5

Which type of graph is most suitable for visualizing the relationship between two continuous variables, like a car's weight and its mileage?

Now that you're able to bring your data to life with visuals, you're well-equipped to explore and present your findings with clarity and impact.

Introduction to Stata