This data set is based on an epidemiological survey of 2484 people to investigate snoring as a risk factor for heart attack. The data set is discussed in Categorical Data Analysis by Agresti (2002). The first variable is “snoring_severity” as reported by the spouses of subjects. The third variable is the number of people with that level of snoring severity, and the second variable is the number of people people in that group who have heart disease. Dividing the second variable by the third one gives you proportion (rate) of people with heart disease for each level of snoring severity.  

snoring_severity heart_disease total
0 24 1379    
2 35 638    
4 21 213    
5 30 254