margins and marginsplot for a binary predictor variable

Stata's margins and marginsplot commands are powerful tools for visualizing the results of regression models. We will use linear regression below, but the same principles and syntax work with nearly all of Stata's regression commands, including probit, logistic, poisson, and others. You will want to review Stata's factor-variable notation if you have not used it before.

Let's begin by opening the nhanes2l dataset. Then let's describe and summarize the variables bpsystol and diabetes.

. webuse nhanes2l
(Second National Health and Nutrition Examination Survey)

. describe bpsystol diabetes


Variable      Storage   Display    Value
    name         type    format    label      Variable label
                                                                                
bpsystol        int     %9.0g                 Systolic blood pressure
diabetes        byte    %12.0g     diabetes   Diabetes status


. summarize bpsystol diabetes


    Variable 
 
        Obs        Mean    Std. dev.       Min        Max
 
 
 
    bpsystol 
 
     10,351    130.8817    23.33265         65        300
    diabetes 
 
     10,349    .0482172    .2142353          0          1

We are going to fit a series of linear regression models for the outcome variable bpsystol, which measures systolic blood pressure (SBP) with a range of 65 to 300 mmHg, and diabetes measures diabetes status with a range of 0 to 1.

margins and marginsplots for a binary variable

Let's fit a linear regression model using the continuous outcome variable bpsystol and the binary predictor variable diabetes. Note that I have used factor-variable notation to tell Stata that diabetes is a categorical predictor.

. regress bpsystol i.diabetes


      Source 
 
       SS           df       MS  
   Number of obs   =    10,349
 
 
 
   F(1, 10347)     =    244.99
       Model 
 
  130296.034         1  130296.034
   Prob > F        =    0.0000
    Residual 
 
  5502984.01    10,347  531.843434
   R-squared       =    0.0231
 
 
 
   Adj R-squared   =    0.0230
       Total 
 
  5633280.05    10,348   544.38346
   Root MSE        =    23.062




    bpsystol 
 
 Coefficient  Std. err.      t    P>|t|     [95% conf. interval]
 
 
 
    diabetes 
 
                                                                
   Diabetic  
 
   16.56328   1.058212    15.65   0.000     14.48898    18.63758
       _cons 
 
    130.088   .2323666   559.84   0.000     129.6325    130.5435

The regression output tells us that the expected SBP for a person without diabetes is 130.088 and the expected SBP for a person diabetes is 16.56328 mmHg higher than a person without diabetes.

We can calculate the expected SBP in each group using the estimated coefficients in the regression output. We substitute a value of 0 in the equation for the nondiabetic group and a value of 1 in the equation for the diabetic group.

. display "E(SBP | no diabetes) = "  130.088 + 16.56328 * 0
E(SBP | no diabetes) = 130.088

. display "E(SBP | diabetes) = "  130.088 + 16.56328 * 1
E(SBP | diabetes) = 146.65128

Stata's margins command will estimate the expected SBP in both groups. Note that the “i.” prefix is required in the regress command but not in the margins command.

. margins diabetes



              
 
            Delta-method                                        
              
 
     Margin   std. err.      t    P>|t|     [95% conf. interval]
 
 
 
     diabetes 
 
                                                                
Not diabetic  
 
    130.088   .2323666   559.84   0.000     129.6325    130.5435
    Diabetic  
 
   146.6513   1.032385   142.05   0.000     144.6276     148.675

The output also reports a standard error, t statistic, p-value, and 95% confidence interval for each estimate. The t statistic tests the null hypothesis that the expected SBP is zero.

We can plot the marginal predictions and their 95% confidence intervals by typing marginsplot.

. marginsplot

Variables that uniquely identify margins: diabetes

By default, marginsplot creates a profile plot using lines. We can use the recast(bar) option if we prefer a bar chart, or “dynamite plunger plot”.

. marginsplot, recast(bar)

Variables that uniquely identify margins: diabetes

We can add the horizontal option to create a horizontal bar chart.

. marginsplot, recast(bar) horizontal

Variables that uniquely identify margins: diabetes

Let's add some additional options to make our graph look nicer. We can use the plotopts(barwidth()) option to add some space between the bars. And we can use the title(), subtitle(), xtitle(), and ytitle() options to add various titles to our graph.

. marginsplot, recast(bar) horizontal plotopts(barwidth(0.8)) 
     title("Expected systolic blood pressure (mmHg)") 
     subtitle("By diabetes status") 
     xtitle("Expected systolic blood pressure (mmHg)") 
     ytitle("")

Variables that uniquely identify margins: diabetes

You can read more about factor-variable notation, margins, and marginsplot in the Stata documentation. You can also watch a demonstration of these commands by clicking on the links to the YouTube videos below.

See it in action

Watch Introduction to margins in Stata, part 1: Categorical variables.

Watch Stata Quick Tip: Margins.

Watch Profile plots and interaction plots in Stata: A single categorical variable.